Go Back   IceInSpace > Beginners Start Here > Beginners Equipment Discussions

Reply
 
Thread Tools Rate Thread
  #1  
Old 18-12-2018, 10:25 AM
Jasp05 (Aaron)
Registered User

Jasp05 is offline
 
Join Date: Nov 2016
Location: Rockhampton
Posts: 226
What specs on a CCD should I be looking at?

I've been doing some research on getting a cooled camera to replace my unmodded DSLR.

I know some basics about the ccd/ cmos cameras but not sure what kind of value/ impact certain parameters will actually make to my imaging.

I know the larger the pixels the more light they collect = shorter exposures.

Quantum efficiency is to do with the amount of light that hits the sensor that gets recorded as signal?? 80% QE is 80% of all photons get recorded?


Full well capacity is the depth of the pixel to record signal before being saturated. Is a small well capacity a bad thing? There can be some quite large differences in this parameter. Example being an ASI1600mm has 20k full well but an asi294MC PRO has nearly 64k. 3 times the capacity..

How much does actual Pixel count affect the images? as in Megapixels. Does more Pixels make the image seem smoother with more detail? (With the understanding that more pixels generally means smaller pixel size and more exposure time required).

And is read noise / dark current a major factor for a beginner? looking at the stats of the ZWO cameras these 2 parameters don't seem to vary a whole lot. somewhere between 1.2 & 2e. So I'm going to say it's not a huge priority?


Budget is a bit tight at the moment.(Around $1k ) But I think I would spend the extra if it was going to make a worthwhile difference. I hear alot of good things about the ASI1600mm PRO but I also see ALOT of them for sale. So not sure what to make of that.

If anyone has any input on the above, or recommendations of what to look for, I'm all ears.
Reply With Quote
  #2  
Old 18-12-2018, 10:32 AM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by Jasp05 View Post
Quantum efficiency is to do with the amount of light that hits the sensor that gets recorded as signal?? 80% QE is 80% of all photons get recorded?
The higher the QE the more sensitive your camera is and it will depend on wavelength so you can find the curves online for your chip model. 80% is very good. Mine is ~50%

Quote:
Originally Posted by Jasp05 View Post
Full well capacity is the depth of the pixel to record signal before being saturated. Is a small well capacity a bad thing? There can be some quite large differences in this parameter. Example being an ASI1600mm has 20k full well but an asi294MC PRO has nearly 64k. 3 times the capacity..
The longer you expose, the more you fill your well to the point you won't record anymore when you reach that limit and it saturates (pure white). In general a camera with deep wells will have a bigger dynamic range for long exposures (from dark to bright) than a camera with shallow well for the same exposure time where data might be white clipped.

Quote:
Originally Posted by Jasp05 View Post
How much does actual Pixel count affect the images? as in Megapixels. Does more Pixels make the image seem smoother with more detail? (With the understanding that more pixels generally means smaller pixel size and more exposure time required).
The smaller the pixel size the finer the details you can capture, provided your scope image scale is matched to your camera.

Don't know much about the ZWO cameras. Read noise becomes an issue when you do a lot of short subs, less when you do very long subs.

For $1k you can pick a very decent second hand camera.
Reply With Quote
  #3  
Old 18-12-2018, 02:03 PM
sil's Avatar
sil (Steve)
Not even a speck of dust

sil is offline
 
Join Date: Jun 2012
Location: Canberra
Posts: 1,474
What do you want to shoot?

The astro imagers tend to be strong for some targets and weak for others. Just like OTAs, it "all depends". For planetary you want frames per second since shooting duration is short before the target rotates and details after stacking get diminished. Solar you also want frames per second since solar features can be so short lived you want to capture as much data to be able to choose the best results from. But megapixels are not needed for planetary but are good for solar. In both cases MP gives you extra detail and with planetary you can record a cropped view around the planet which can gain you many frames per second increase so its nice there. Megapixels are good too for most deep space targets to maximise the number of pixels per arcsecond possible since only some targets are going to fill the sensor field and after stacking etc you are probably going to be cropping anyway. but megapixels is more a "nice to have" feature.

Must have: USB3, exposure speeds suitable to what you want to do and image with it, a dedicated (not shared with OS) large SSD drive is essential for frames per second performance and zero dropped frames. over 150fps is easy here for planetary/solar/lunar. File sizes will be huge too. I record to SSD and transfer to portable HDD to take to another machine to store on my nas where i can access for processing. Keep recording machine JUST for recording, no processing, no games, no web. everything you do with it will start to bloat and slow it down. no internet access needed ever. Setup control and guiding should be fine though.

When you sort out the zwo cameras according to their specs you find they have high speed planetary cams and deep space oriented ones, and everything in between. If you want "the one that does everything the best" well like OTAs is doesn't exist. Its all about compromises somewhere but there are some good ones out there. I think my current one is the 178MC-C which is for planetary really but it has capability for deep space long exposure stuff just not as well as others which in turn will be poor for planetary. You're on the right track with the homework you've been doing, comprehend whats needed for types of targets ideally and take note of the spec that are relevant to that and i suggest you go through say all the zwo cams and note them down with specs, which I did and it was very quick to be able to rule out various cams, eg non cooled just ignore, usb2 only ignore, mono ignore, low fps capture ignore and was left with a handful of planetary contenders for me which was easy to go through each of them for other specs that were either undesirable for my immediate need or had desirable features for a future need.

You'll also find competing makers have similar range in their models as they all use the same sensor mostly at specific price/feature points. Its not as daunting a task as you think when setting out to pick a cam. Also keep in mind the colour vs mono capabilities. I havent delved into mono capture yet and filter wheels, but I kept it in mind that i wanted a cam for a specific need now and one that hopefully wont be useless when i start to explore mono capture.
Reply With Quote
  #4  
Old 18-12-2018, 04:09 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Quote:
Originally Posted by sil View Post
What do you want to shoot?
That's definitely the #1 question you need to answer. Starting with planetary vs deep sky. If the choice is deep sky then you need to think about what sort of objects, field of view and image scale you need.

Quote:
Originally Posted by multiweb View Post
The higher the QE the more sensitive your camera is and it will depend on wavelength so you can find the curves online for your chip model. 80% is very good. Mine is ~50%
Just one point of caution on QE: some manufacturers supply a graph showing absolute QE which is what Marc is talking about above; other manufacturers supply a graph showing relative QE. A relative QE graph will show you how the sensitivity of the camera varies with wavelength but won't allow you to compare actual QE with a different camera.

Quote:
Originally Posted by Jasp05 View Post
Full well capacity is the depth of the pixel to record signal before being saturated. Is a small well capacity a bad thing? There can be some quite large differences in this parameter. Example being an ASI1600mm has 20k full well but an asi294MC PRO has nearly 64k. 3 times the capacity..
Deep wells are useful with a camera that has high read noise when you'll want to do long exposures. Shallow wells are fine with a low read noise camera and short subs, and this combination is less demanding on your guiding.

Quote:
Originally Posted by Jasp05 View Post
How much does actual Pixel count affect the images? as in Megapixels. Does more Pixels make the image seem smoother with more detail? (With the understanding that more pixels generally means smaller pixel size and more exposure time required).
More pixels will give you an image that looks better at larger sizes and also provides the option to crop and still have something left. As you say, this will come at the cost of slower imaging.

When you think about pixel count and sensor size you need to also take the scope into consideration. If you have long focal length then larger pixels will be a better match. For a short focal length refractor then smaller pixels will give you more detail. This needs to be considered in terms of your overall imaging goals (type of objects, desired field of view, desired level of detail = image scale.)

Quote:
Originally Posted by Jasp05 View Post
And is read noise / dark current a major factor for a beginner? looking at the stats of the ZWO cameras these 2 parameters don't seem to vary a whole lot. somewhere between 1.2 & 2e. So I'm going to say it's not a huge priority?
The length of time for individual subs of good quality increases with the square of the read noise, so it is a consideration especially if you want to do short subs. CCDs typically have much higher read noise than the CMOS cameras you have been looking at. Note that even CMOS cameras can have high read noise if you run them at low gain settings.

Quote:
Originally Posted by Jasp05 View Post
Budget is a bit tight at the moment.(Around $1k ) But I think I would spend the extra if it was going to make a worthwhile difference. I hear alot of good things about the ASI1600mm PRO but I also see ALOT of them for sale. So not sure what to make of that.
The ASI1600mm is a good camera if it suits your needs. I'm using one and the only thing I'd like is a much bigger version of the sensor I wouldn't use it on one of my long focal length scopes but for camera lenses and short focal length refractors it is a great fit.

I'd recommend trawling around Astrobin taking a look at pics taken with some of the cameras you are interested in. Take note of the other gear used and the amount of integration time. That might give you some ideas...

Cheers,
Rick.
Reply With Quote
  #5  
Old 18-12-2018, 09:33 PM
Jasp05 (Aaron)
Registered User

Jasp05 is offline
 
Join Date: Nov 2016
Location: Rockhampton
Posts: 226
Well just to clear a few things up. I am only interested in Deep space imaging atm. Planetary I've already got another camera that I could use it for it. (It's a cheapo one that was given to me when I bought a dobsonian from a guy. But it works).


My current scope is a Skywatcher ED72 with 0.85x reducer. So Focal length of 357mm. As I'm still new to the hobby my focus is widefield nebula, and the larger galaxies.

I don't see myself upgrading anytime soon from my current setup. (Too much to do at this focal length and learn before I go to long FL scopes).

I do plan to image from my backyard. Apparently is around a Bortle 5 sky based on Light pollution maps.

I have considered Mono due to being able to deal with light pollution a bit better. But am also considering one shot colour and a light pollution filter. (If anyone can comment on imaging from a bortle 5 sky and whether a mono would really be required please do).

When you say short subs, how long are you talking? is 2-5min short?

And as far as matching my scopes FL to the camera pixel size how do you do that calculation?

I've looked on astronomy tools using the ccd calculators and it tells me a pixel size of less than 3um is ideal for my setup for poor to ok seeing. and under 2um for good or better seeing.

My DSLR pixel size is 4.3um.

Based on the above an ASI183mm-C is a 20.1 mega pixel, 2.4um sensor with 15k well depth and QE of 84%. Read noise is 1.6e which is low.

So that would be a better camera for my setup than the ASI1600mm which is 16 megapixel, 3.8um, with 20k well depth and 1.2e read noise.

Which would you take out of those 2 cameras given my setup and why?
Reply With Quote
  #6  
Old 19-12-2018, 07:57 AM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Either camera would be a reasonable choice.

The ASI1600mm would have an image scale of 2.2 arcsec/pixel and a FOV of 2.84x2.14 degrees.

The ASI183mc would have an image scale of 1.39 arcsec/pixel and a FOV of 2.12x1.42 degrees.

(I have a spreadsheet but you can find formulae for determining image scale etc at http://www.wilmslowastro.com/software/formulae.htm)

ASI1600mm thoughts: wider field and lower resolution, more expensive but has the option of doing narrow band in future, mono processing arguably has a slightly steeper learning curve, image scale will be forgiving of bad seeing and less than perfect guiding. Will be fast enough to do short subs (maybe 30 to 120 secs?)

ASI183mc thoughts: smaller field and higher resolution, image scale still OK for typical seeing and not too demanding on guiding. Will be quite a lot slower than the ASI1600mm due to smaller pixel size and lower QE (84% is peak mono QE but you're imaging through a Bayer matrix and only getting 1/3 of the photons in each pixel.) Will still be OK for shortish subs due to low read noise but not as short as ASI1600mm.

My spreadsheet suggests that the ASI1600mm will collect luminance about 5 times faster than the ASI183mc. The difference is less for colour filters... more like 2 times.

Cheers,
Rick.
Reply With Quote
  #7  
Old 19-12-2018, 09:48 AM
Jasp05 (Aaron)
Registered User

Jasp05 is offline
 
Join Date: Nov 2016
Location: Rockhampton
Posts: 226
Quote:
Originally Posted by RickS View Post
Either camera would be a reasonable choice.

The ASI1600mm would have an image scale of 2.2 arcsec/pixel and a FOV of 2.84x2.14 degrees.

The ASI183mc would have an image scale of 1.39 arcsec/pixel and a FOV of 2.12x1.42 degrees.

(I have a spreadsheet but you can find formulae for determining image scale etc at http://www.wilmslowastro.com/software/formulae.htm)

ASI1600mm thoughts: wider field and lower resolution, more expensive but has the option of doing narrow band in future, mono processing arguably has a slightly steeper learning curve, image scale will be forgiving of bad seeing and less than perfect guiding. Will be fast enough to do short subs (maybe 30 to 120 secs?)

ASI183mc thoughts: smaller field and higher resolution, image scale still OK for typical seeing and not too demanding on guiding. Will be quite a lot slower than the ASI1600mm due to smaller pixel size and lower QE (84% is peak mono QE but you're imaging through a Bayer matrix and only getting 1/3 of the photons in each pixel.) Will still be OK for shortish subs due to low read noise but not as short as ASI1600mm.

My spreadsheet suggests that the ASI1600mm will collect luminance about 5 times faster than the ASI183mc. The difference is less for colour filters... more like 2 times.

Cheers,
Rick.
Thanks Rick. I must admit I'd be interested in having a peek at your spreadsheet if you wouldn't mind sharing it.

Would definately like to know how you estimate how long it would take to collect that luminance / colour data between cameras.
Reply With Quote
  #8  
Old 19-12-2018, 10:22 AM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Hi Aaron,

Here's a spreadsheet with a few example systems. The model is simple and doesn't take all factors into account, but it does give a rough relative idea of how combinations of different scopes and cameras will perform.

Cheers,
Rick.
Attached Files
File Type: zip Astrograph efficiency - IIS.zip (11.1 KB, 13 views)
Reply With Quote
  #9  
Old 19-12-2018, 11:47 AM
Jasp05 (Aaron)
Registered User

Jasp05 is offline
 
Join Date: Nov 2016
Location: Rockhampton
Posts: 226
Thanks again Rick,

Where do you get the "Signal" Value from?

Why does SNR = Square root of signal?
Why does Time = 1/ signal?

Sorry for all the questions, but I do like to get a handle on the fundamentals of these things.
Reply With Quote
  #10  
Old 19-12-2018, 01:36 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Quote:
Originally Posted by Jasp05 View Post
Thanks again Rick,
Where do you get the "Signal" Value from?
The rate at which signal is captured is proportional to the square of the aperture, the square of the pixel size (both measures of area) times an efficiency factor and inversely proportional to the square of the focal length. The efficiency factor takes into account central obstruction (if any), optical efficiency (100% less losses from from refraction and reflection) and QE.

I'm not trying to calculate in specific units. It's just a relative measure for comparison purposes.

Quote:
Originally Posted by Jasp05 View Post
Why does SNR = Square root of signal?
The major source of noise should be shot noise (assuming your subs are long enough to be sky limited) and the amount of shot noise is the square root of the signal, hence signal/noise is also the square root of the signal.

Quote:
Originally Posted by Jasp05 View Post
Why does Time = 1/ signal?
Time is again a relative measure representing how long it takes to accumulate signal. If you're collecting 10 e-/second then the inverse is 1/10 sec/e-.
Reply With Quote
  #11  
Old 19-12-2018, 02:49 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
What's the Eff? (no pun intended) Strehl ratio?
Reply With Quote
  #12  
Old 19-12-2018, 03:05 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Quote:
Originally Posted by multiweb View Post
What's the Eff? (no pun intended) Strehl ratio?
For the optical component I think I assumed 0.995% transmission for a lens and 0.97% reflectivity for a mirror. Optical quality is not considered.
Reply With Quote
  #13  
Old 19-12-2018, 03:12 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by RickS View Post
For the optical component I think I assumed 0.995% transmission for a lens and 0.97% reflectivity for a mirror. Optical quality is not considered.
You've got 0.927 for the C300 and 0.980 for the refractors.
So for a FSQ I'd use 0.980 or 0.995?
For the CN-212, 0.927 or 0.980?
Reply With Quote
  #14  
Old 19-12-2018, 03:23 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Quote:
Originally Posted by multiweb View Post
You've got 0.927 for the C300 and 0.980 for the refractors.
So for a FSQ I'd use 0.980 or 0.995?
For the CN-212, 0.927 or 0.980?
0.98 is about right for a FSQ/4-lens system.
0.927 was calculated for two mirrors and a 3 lens corrector.

I don't obsess too much about the YMMV factors

Cheers,
Rick.
Reply With Quote
  #15  
Old 19-12-2018, 03:48 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Cool

Last question - QE(Ha). For the QHY9 the curve with microlens says ~45% @ 650nm so I enter 0.45 in that cell right?

Then I think I have everything else working.

PS: central obstruction by area or diameter?
Reply With Quote
  #16  
Old 19-12-2018, 03:57 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,584
Quote:
Originally Posted by multiweb View Post
Last question - QE(Ha). For the QHY9 the curve with microlens says ~45% @ 650nm so I enter 0.45 in that cell right?
Yep. No special reason why I used the QE for Ha except that I seem to capture it frequently It would also make sense to put in an average QE over the visible range but it's tougher to estimate.

Quote:
Originally Posted by multiweb View Post
PS: central obstruction by area or diameter?
By diameter like what is normally quoted in specs, e.g. 45% obstruction would be 0.45.
Reply With Quote
  #17  
Old 19-12-2018, 04:01 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by RickS View Post
Yep. No special reason why I used the QE for Ha except that I seem to capture it frequently It would also make sense to put in an average QE over the visible range but it's tougher to estimate.
Cool.

Quote:
Originally Posted by RickS View Post
By diameter like what is normally quoted in specs, e.g. 45% obstruction would be 0.45.
Yeah I'm asking because I checked the specs on the C11 here and it quoted 34% by diameter and 12% by area. That's a big difference.
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 08:06 PM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement