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

Selecting the Best Camera Pixel Size

astrophotography
  • Please log in to reply
10 replies to this topic

#1 BobinBend

BobinBend

    Vostok 1

  • -----
  • topic starter
  • Posts: 173
  • Joined: 02 Mar 2018
  • Loc: Bend, Oregon

Posted 19 February 2019 - 02:51 PM

I need some help to better understand the considerations in purchasing the optimum CCD OSC camera for my current gear. I currently have a Celestron 8" SCT with a 1280mm focal length (I'm using a focal reducer) and a Focal Ratio of F6.3. My primary camera is an unmodified Canon 60D which has a Pixel Size of 4.31microns. I would like to purchase a CCD OSC camera that is ideally cooled but I'm not sure what characteristics I should be looking at for an ideal match.

 

Any suggestions would be appreciated.

Thanks,

Bob



#2 Michael Harris

Michael Harris

    Vostok 1

  • -----
  • Posts: 195
  • Joined: 07 Mar 2010
  • Loc: Kentucky, USA

Posted 19 February 2019 - 03:00 PM

Try the site astronomy.tools. They have a page that will tell you how various camera choices will work with that scope, based on your “average” seeing. 

 

https://astronomy.to...ccd_suitability


  • Gary Z likes this

#3 sg6

sg6

    Aurora

  • *****
  • Posts: 4595
  • Joined: 14 Feb 2010
  • Loc: Norfolk, UK.

Posted 19 February 2019 - 03:04 PM

The method is that you determine the size of say a 1 arcsec or 2 arcsec object then pick a pixel size that matches.

That is about the easiest description.

 

So determine how big 1 arcsec covers with your focal length.

2 arc sec is so close that you can simply double the size.

 

Say the 1 arcsec is 5um, then you are likely needing to look at a pixel size of 5um to 10um. 10 is I guess "big" so 5 to 7 seems appropriate.

 

My value of 1um comes from an artical by SBIG, also 1 is easy to remember. They said 1 arcsec and others say 2 arcsec so work out both and try the middle.

 

Like all astro stuff there are exceptions and imaging planets are the exception where you can go smaller, say 0.5 arcsec. But remember that the majority of images will not be planets. 3 planets and 110 Messiers.



#4 DuncanM

DuncanM

    Vanguard

  • -----
  • Posts: 2348
  • Joined: 03 Nov 2009
  • Loc: Arizona Sky Village or the rain forest

Posted 19 February 2019 - 03:23 PM

CCD Calculator is a handy tool:

 

http://www.newastro....camera_app.html (be aware that some of the cameras in it's database have incorrect info)

 

and you can add new cameras and scopes easily.

 

The Celestron F6.3 reducer has a rather small area of good correction. You might want to try some images via your 60D to see what the largest image circle is acceptable to you and if you decide to retain it, then you might want a camera with a smaller sensor than on the 60D.


Edited by DuncanM, 20 February 2019 - 10:55 AM.


#5 pyrasanth

pyrasanth

    Apollo

  • *****
  • Posts: 1429
  • Joined: 08 Jan 2016

Posted 19 February 2019 - 03:45 PM

The problem I see a lot and I guess it affects other people is simply the quality of the atmosphere more commonly called seeing. We have been lured by small pixel sizes promising potential resolutions to the limits of our optics. Sadly, the reality, is somewhat different. I would be very lucky where I live to get under 0.7 arc secs per pixel and more realistic is just under 1 arc sec.

 

Using the C14 I'm going to be binning my Moravian G3 16200 to 2x2 and forgetting 1x1 until I can see a night of exceptional seeing. If my stars are hopping on the guider like a frog in a pond then I know this will be the right decision. Until I move my telescope to the moon 2x2 bin is about right on the C14 for my seeing conditions. That is the one important thing you need to determine for your camera and focal length balanced with the quality of your localities seeing conditions.

 

Your telescope focal length and seeing conditions will & should strongly influence your choice of camera!


Edited by pyrasanth, 19 February 2019 - 03:52 PM.

  • ks__observer likes this

#6 pyrasanth

pyrasanth

    Apollo

  • *****
  • Posts: 1429
  • Joined: 08 Jan 2016

Posted 19 February 2019 - 03:50 PM

The method is that you determine the size of say a 1 arcsec or 2 arcsec object then pick a pixel size that matches.

That is about the easiest description.

 

So determine how big 1 arcsec covers with your focal length.

2 arc sec is so close that you can simply double the size.

 

Say the 1 arcsec is 5um, then you are likely needing to look at a pixel size of 5um to 10um. 10 is I guess "big" so 5 to 7 seems appropriate.

 

My value of 1um comes from an artical by SBIG, also 1 is easy to remember. They said 1 arcsec and others say 2 arcsec so work out both and try the middle.

 

Like all astro stuff there are exceptions and imaging planets are the exception where you can go smaller, say 0.5 arcsec. But remember that the majority of images will not be planets. 3 planets and 110 Messiers.

10 um is only big if you have short focal lengths like a very fast instrument like the Celestron RASA. If you use an SCT then 10 um is going to be pretty much the perfect size especially when you get to the C11 class and above with focal lengths over 3 metres.



#7 DuncanM

DuncanM

    Vanguard

  • -----
  • Posts: 2348
  • Joined: 03 Nov 2009
  • Loc: Arizona Sky Village or the rain forest

Posted 19 February 2019 - 04:09 PM

The problem I see a lot and I guess it affects other people is simply the quality of the atmosphere more commonly called seeing. We have been lured by small pixel sizes promising potential resolutions to the limits of our optics. Sadly, the reality, is somewhat different. I would be very lucky where I live to get under 0.7 arc secs per pixel and more realistic is just under 1 arc sec.

 

Using the C14 I'm going to be binning my Moravian G3 16200 to 2x2 and forgetting 1x1 until I can see a night of exceptional seeing. If my stars are hopping on the guider like a frog in a pond then I know this will be the right decision. Until I move my telescope to the moon 2x2 bin is about right on the C14 for my seeing conditions. That is the one important thing you need to determine for your camera and focal length balanced with the quality of your localities seeing conditions.

 

Your telescope focal length and seeing conditions will & should strongly influence your choice of camera!

Generally speaking, you want your pixel size in arc seconds to be 1/2, or less, then your seeing. This will allow deconvolution via software to restore some of the lost resolution.  If you are using a C-14 at F11 (4000mm EFL) then 2 x 2 binning will result in ~.9 arc/secs /pixel and would be suitable for ~2 arc sec or worse seeing. However, if you are using the FR and imaging at ~F7.5 (2500mm EFL) then 2x2 binning might result in lost resolution if you decided to use deconvolution.

 

EDIT: I see that the KAF16200 uses 6 micron pixels, not 9 microns as per my calculations above.


Edited by DuncanM, 19 February 2019 - 04:46 PM.

  • Jon Rista and pyrasanth like this

#8 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 22337
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 19 February 2019 - 04:40 PM

The problem I see a lot and I guess it affects other people is simply the quality of the atmosphere more commonly called seeing. We have been lured by small pixel sizes promising potential resolutions to the limits of our optics. Sadly, the reality, is somewhat different. I would be very lucky where I live to get under 0.7 arc secs per pixel and more realistic is just under 1 arc sec.

 

Using the C14 I'm going to be binning my Moravian G3 16200 to 2x2 and forgetting 1x1 until I can see a night of exceptional seeing. If my stars are hopping on the guider like a frog in a pond then I know this will be the right decision. Until I move my telescope to the moon 2x2 bin is about right on the C14 for my seeing conditions. That is the one important thing you need to determine for your camera and focal length balanced with the quality of your localities seeing conditions.

 

Your telescope focal length and seeing conditions will & should strongly influence your choice of camera!

You are using a very long focal length scope there with the C14. It's about image scale in the end. With 6 micron pixels at 3900mm, your image scale is about 0.32"/px, which is a very, very large scale (high magnification, low flux per pixel). IMO, based on my own experience and data I've worked from other people, the practical limit for most skies, even given very good seeing of say 1", is about 0.5"/px image scale. Beyond that, you don't really gain much, and you certainly lose something. Even at 0.5"/px, you may be losing too much, and the practical limit for less advanced imagers may be between 0.6-0.8"/px. 

 

But, tiny pixels do offer something, and this is something I've benefitted from myself. With 4, 3, 2 micron pixels, you can achieve these large image scales more easily, with less expensive equipment. THAT is the lure. Now, it would be tough to beat a 14" SCT with 9-15 micron pixels...the big aperture trumps there, once you have sufficient image scale to make ideal use of it. However, the performance of say an 11" SCT with a KAF-16200, could be achieved with a 10" f/4 Newt and an IMX183. You would have roughly the same image scale, higher Q.E. and actually a slightly larger field of view. Roughly the same aperture, so you would be gathering about the same light (slight differences in CO, a little smaller with the newt, and the slight difference in aperture would be offset by the high Q.E. of the IMX183 sensor).

 

You can spend a lot less, and still be able to do high resolution imaging. It isn't just about the camera, either...a big camera and a big, heavy scope like that requires a more heavy duty mount, so the total cost difference is potentially in the tens of thousands of dollars range.

 

This only really matters if high resolution imaging matters to you. For wide field imaging, image scale is less important, and oversampling by a sufficient amount is not as important as getting the right size field. A smaller image scale/lower magnification of the object allows for higher performance imaging as well, and when scopes (apertures) get smaller, diffraction grows and can often become a larger blur term than seeing itself, and that limits how small of pixels you ultimately need to get good resolution.


  • pyrasanth likes this

#9 pyrasanth

pyrasanth

    Apollo

  • *****
  • Posts: 1429
  • Joined: 08 Jan 2016

Posted 19 February 2019 - 05:10 PM

Hi Jon,

 

I see all the arguments which make sense. I use the C14 with the reducer so my image scale unbinned with the 16200 is about 0.45 arc secs per pixel which for my seeing conditions is most certainly over sampled hence the move to 2x2 binning which gives a more realistic 0.9 arc secs per pixel. There is nothing stopping me going 1x1 bin for the L and capitalising on the lower spatial resolution for the colour at 2x2 bin. That should capture a bit of both worldly benefits.

 

As a side note remember I have the 11" RASA which I need to put on its own mount in a few months. That when coupled with a high res CMOS camera will give me the best of both worlds so I will be set for both camps.


Edited by pyrasanth, 19 February 2019 - 05:22 PM.


#10 OldManSky

OldManSky

    Ranger 4

  • -----
  • Posts: 385
  • Joined: 03 Jan 2019
  • Loc: Valley Center, CA USA

Posted 19 February 2019 - 05:16 PM

I used to have custom license plates on my astro-suv that were "1ARCSEC".

When people asked me what that was, I'd tell them it's something I try for but rarely get when doing astrophotography :)


  • Ken Sturrock likes this

#11 BobinBend

BobinBend

    Vostok 1

  • -----
  • topic starter
  • Posts: 173
  • Joined: 02 Mar 2018
  • Loc: Bend, Oregon

Posted 20 February 2019 - 10:40 AM

Thanks to all for your input and suggestions. I have a much better idea of what to look for and I appreciate all the additional links and tools to help me get there.

 

Bob




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





Also tagged with one or more of these keywords: astrophotography



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics