Go Back   IceInSpace > Equipment > Equipment Discussions
Register FAQ Calendar Today's Posts Search

Reply
 
Thread Tools Rating: Thread Rating: 2 votes, 3.00 average.
  #1  
Old 04-05-2011, 04:28 PM
roughy (Mark)
Registered User

roughy is offline
 
Join Date: May 2011
Location: Hobart
Posts: 134
What to look for in a CCD

Hi. I've been an avid visual astronomer for many years and have decided to move into imaging.

I would like to purchase a reasonable quality camera that will suit most purposes for a long time to come. For example, I think the Atik 383 or SBIG 8300 (both around 8mp) seem to fit this bill. Their price range is also my preferred option. However my knowledge of this topic is extremely limited. Some help please.

Why do monochrome and filters seem to be preferred to one shot colour?
Is it better to consider the relationship between pixel size and array size rather than getting the biggest array?
Are there just as good 2 and 4 megapixel options even though some of these appear to be a bit more expensive?

I am about to set up an RC8" and will also be getting a 100 to 110mm refractor in the medium to long term.

Any advice would be greatly appreciated. Many thanks.

Mark
Reply With Quote
  #2  
Old 04-05-2011, 05:47 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 18,183
Hi Mark,

I'll attempt to answer that.

First consideration should be your local seeing conditions and the focal length of your scope. The camera needs to be matched to the scope and seeing conditions.

Here is a free calculator which can work this out for you with different scope and camera combinations (the camera list is out of date but you can plug in the numbers):

http://new-astronomy-ccdcalc.software.informer.com/

What this means is for an RC8 of 203mm aperture and 1624mm focal length an 8300 chipped camera will give .68 arcseconds per pixel.

The 8300 chip has 5.4 micron pixels (smallish) and 3326 x 2504 in the array (or a number very similar).

Assuming average seeing of around 2-3 arcseconds (an arc second =1/60th of a degree ie. a pretty small angle) and assuming sampling theory is correct which states you want a minimum of 2X perhaps 3X times to get an accurate sample. Then you get 3 arcsecond seeing divided by 3 = 1 arcsecond or in better seeing 2 arc seconds divided by 3 = .66 arc seconds.

So .68 is close to ideal sampling. This means then your camera will produce maximum sensitivity and resolution and that the light is not being spread too thin over the chip or not too broadly.

With a refractor of 115mm or so, typically that will mean about 800-900mm focal length so you will get 1.21 arc seconds per pixel again not too far away from the ideal.

The smaller pixels of the 8300 tend to work better with faster scopes.

Now as far as which 8300 chipped camera to get then here are the considerations:

1. Cost.
2. Performance (they are not all the same by a long ways)
3. Accessories available
4. Autoguiding solutions available.

1. is obvious but there are some hidden costs. Say the SBIG ST8300, very reasonably priced and a popular camera. But it needs a filter wheel, filters and an autoguiding solution (no autoguider or dual chip system).

So make sure you factor in the cost of the camera, the cost of a filterwheel (some have filter wheels built in - well really only the QSI)
and autoguiding.

2. Performance:

Not all 8300 chipped cameras have the same performance.
Performance can be broken down into - sensitivity (called QE for quantum efficiency or how efficiently does the chip convert a photon into an electron). You get a bit more QE with no cover slip over the chip. A FLI Microline 8300 or an Apogee U8300 can be specified with no cover slip. This can also reduce annoying small halos around stars.
Without the cover slip a FLI Microline 8300 gives 60% QE which is very high.

Cooling. Cooling reduces noise. Higher cooling then is better than lesser cooling. The best cooling cameras I am aware of are the FLI and the Apogee both slam the 8300 down to -35C easily. The FLI does it very quickly the Apogee very slowly (30 minutes) which some don't like.
ST8300 not sure of but probably 40C below ambient which will give you
-20C most of the time and -25 to -30C in winter. That's quite cool and the chip should be quite clean at that temp.

QSI is the weakest in cooling although I see they have a series 600 with enhanced cooling now to address that weakness.

Next is readout noise. This is the noise the camera generates during the chip reading process. FLI is lowest. Apogee is very low. Not sure where the SBIG is there. It all adds up.

Next would be miscellaneous noise from the electronics. FLI is very clean
(probably the cleanest) Apogee would not be far behind. Now this is where the SBIG probably falls behind. Its not the end of the world as it cleans up with standard processing but it puts more pressure on that being done very well.

3. Accessories:

Filter wheels, compatibility with electronic focusers, ease of getting adapters, adaptive optics unit availability.

QSI are the best here with a combined filterwheel and offaxis guider giving excellent flex free autoguiding so if you have a decent mount you are very likely to get round stars. Which is a large part of the battle with imaging.

So I know FLI, Apogee and know about QSI and ST8300. You would do well with any of these. QSI is probably best bang for your buck as it incorporates the filter wheel and offaixs guider. FLI is the highest quality and highest performing camera. It is the fastest in download time, it has the best cooling, the lowest noise, has a sealed chamber with inert gas so no desiccant and has a no cover slip for the chip option. It is light, extremely well made and rugged. It has excellent jacks for power and USB (the SBIG has very cheap and weak jacks that can fail).

FLI has RBI control. This means residual bulk images or ghost images. That is if you take an image of something bright these chips tend to keep a ghost of that image into the next image. I think Apogee also may have RBI control as well now.

There is also a new Apogee range called Ascent. New and untried but probably quite good and cheaper.

I can't comment on QHY or Atik except that QHY would probably be a lower scale option and would not perform as well as the above. Atik is an up and coming maker as well but both of these guys were small and cheap type cameras now offering more expensive and bigger chips.

Starlight Express is another excellent camera maker and they have a range of options from guiders to adaptive optics. They have really low noise cameras.

Mono versus Colour. Colour is simpler and requires no filter wheel and filters and is therefore cheaper. Results can be great on the brighter objects but dimmer objects starts to show the lack of sensitivity. Typically mono chips are 50% more sensitive - ie 8300 mono is around 60% QE but one shot colour is more like 30-35% QE. A big drop. You'll see a lot of extra noise in your images especially in the dim areas.

But its a good way to start off with CCD imaging. And its cheaper and every shot counts (you need luminance, red, green and blue filtered images with mono cameras to get a colour shot).

Also mono means you can do narrowband imaging more easily - Halpha and others. One shot colour can do them but not that easily and you lose a lot of performance in that area and they are not really suitable for that unless again its a really bright object.

So QSI is probably the least expensive and most integrated package in the 8300 range as it has an internal filter wheel, and an offaxis guider and if you get the later souped up cooling version decent cooling as well.

That would be my pick.

http://www.optcorp.com/ProductList.a...-320-1232-1876

If you want absolute best then its FLI. Best cooling, highest QE, RBI control, fastest downloads, lowest noise. lots of accessories, no cover slip option.

Next is Apogee (similar to FLI but the slow cooling whilst it doesn't sound like much is very irritating and inconvenient perhaps their new Ascent camera does not do that) and Starlight Express. But these all need filter wheels and autoguiders and off axis guider or a guide scope (not recommended especially for the RC8 you will have hell trying to get round stars).

The 4020 chip is also an excellent chip for imaging you should consider.

Greg.
Reply With Quote
  #3  
Old 04-05-2011, 10:07 PM
pvelez's Avatar
pvelez (Pete)
Registered User

pvelez is offline
 
Join Date: Aug 2006
Location: Sydney
Posts: 1,250
Greg

Where were you 18 months ago when I bought my first cooled ccd? That is an awesome post - thanks for all your effort. I'm sure many will bookmark that one.

I have just bought the QSI - a trade up from the ST8300 I hope. I agree re the guiding issues with the RC8. I had a f4 SN which was pretty forgiving. The change to the RC8 was a bit of a challenge. It can be done but you need to get your polar alignment right.

I agree that the 8300 is a cracker when paired with the RC8. It gives a nice field of view and allows binning if the seeing is less than ideal.

Pete
Reply With Quote
  #4  
Old 05-05-2011, 12:13 AM
Paul Haese's Avatar
Paul Haese
Registered User

Paul Haese is offline
 
Join Date: Jan 2009
Location: Adelaide
Posts: 9,991
Greg, I think that the QSI has RBI control too. This function lights up when I connect my camera and I do not get ghosting on bright objects. I cannot find anything in the online data at QSI but it seems to be the case. Read noise of the QSI is really pretty low too.
Reply With Quote
  #5  
Old 05-05-2011, 12:04 PM
wasyoungonce's Avatar
wasyoungonce (Brendan)
Certified Village Idiot

wasyoungonce is offline
 
Join Date: Jul 2006
Location: Mexico city (Melb), Australia
Posts: 2,359
Good Post Greg.
Reply With Quote
  #6  
Old 05-05-2011, 07:51 PM
higginsdj's Avatar
higginsdj
A Lazy Astronomer

higginsdj is offline
 
Join Date: Mar 2006
Location: Canberra
Posts: 614
Since when is average seeing as good as 2-3 arcseconds? Its certainly not that at HHO. A good rule of thumb is 1-2"/px at least for astrometry/photometry work.

Cheers
Reply With Quote
  #7  
Old 06-05-2011, 09:09 PM
allan gould's Avatar
allan gould
Registered User

allan gould is offline
 
Join Date: May 2005
Location: Brisbane
Posts: 4,485
Excellent unbiased summary Greg! Agree with everything posted and it should be made a sticky somewhere for others to read.
Reply With Quote
  #8  
Old 06-05-2011, 11:12 PM
CoolhandJo's Avatar
CoolhandJo (Paul)
Registered User

CoolhandJo is offline
 
Join Date: Oct 2008
Location: Sydney
Posts: 1,809
Great contribution to the forum! (My Atik cheap! . Well yes but still very effective in tems of QE, cooling, affordability)

Agree with your post when considering top end. Mid range... well?
Reply With Quote
  #9  
Old 07-05-2011, 02:39 PM
marki's Avatar
marki
Waiting for next electron

marki is offline
 
Join Date: Mar 2008
Posts: 2,427
I am suprised the QHY9 with the KAF8300 chip has not made an appearance here. The camera is capable of cooling the the chip 50 degrees C from ambiant, has a very low read noise due to the power board not being on the camera and comes in a package which includes filterwheel, parfocal LRGB filters and camera for less then the cost of any of the competing cameras alone. Members on this site have tested the QHY9 along with the ST8300 and other cameras and sold the others keeping the QHY9. Wonder if that says anything? I have had mine for 18 months and it is a great little camera that is small in size and weight.

Mark
Reply With Quote
  #10  
Old 12-05-2011, 08:28 PM
roughy (Mark)
Registered User

roughy is offline
 
Join Date: May 2011
Location: Hobart
Posts: 134
Many thanks to all for their thoughts and advice. New toy has been ordered - QSI583wsg.
Reply With Quote
  #11  
Old 14-05-2011, 04:53 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 18,183
One other point about CCD cameras that is worth mentioning.

Most of the modern cameras have their firmware on an internal flash memory. The SBI STX does but the ST series and STL series does not. I think Starlight express does as well and not sure about Atik or QHY.

What that means is the drivers are loaded into the camera from the computer and it takes a little while to do that. If the power gets interrupted (which happens very often with my ST402 as it has a cheap and weak power jack - this reminds me I need to replace that jack) then
you will get an error message when you go to take an image.

Reconnecting the power will not reconnect it in the software. You have to close down CCDsoft and reopen after you have repowered the camera. Wait the 25 seconds or so for it to load the drivers and then reconnect. This means losing your cooling etc. and autoguiding and restarting everything. OK, if everything is working well it probably would not happen that often but it is surprising how often it can still occur.

Modern cameras have the drivers in the camera. My Apogee and FLI cameras and Starlight Express cameras connect instantly. The SBIG takes about 20 odd seconds and sometimes fails. The others almost never. Also you can unplug the Apogee and FLI camera and plug it back into the computer and it is still running at the same temperature and ready to go. NICE. Especially at 3am when you are tired and cold. A lot less swearing!

The other consideration is download times. Slow download times mean the autoguider stops when the download is in progress and the tracking errors build up. If you started another exposure straight away you would be starting at maximum guide errors which then reduce over the first 20-30 seconds of the image. So you have to program in a delay before the next imag starts. FLI was fastest with 1 second downloads with a 1x1 binned image - lightning. Apogee wasn't too bad. I see QSI series 600 has the same download speeds now as FLI - excellent. Also they have -45C ambient cooldown. Plus with the fast downloads, the drivers on board, the off axis guiders, the internal filter wheel means I think they have now overtaken FLI in practicality. FLI have some other nice features not sure QSI matches like antireflection coatings on quartz CCD windows, sealed chambers with inert gas that do not need desiccants etc and the option of no cover slip which gains a bit of extra QE and less halos.

The only thing they haven't gotten now is ghost image control. That is called RBI (residual bulk image). That means if you shoot flats at dusk then the chip will retain some of the charge in the deep layers of the silicone. It slowly leaks out during the later exposures meaning your darks will not match the lights 100%. All full frame KAF series chips do this. Some worse than others. The 09000 chip I believe is the worst. Without RBI control you would have great diffifculty in using a 09000 chip.

I have used several KAF chips and I can't say that this has affected my images or if it has I am unaware of its effects. My FLI camera has the ability to handle this by an infrared flash of the sensor before it takes a light. You take your darks also preflashed in IR and you can control how much it is preflashed. Now your darks and lights are matched more exactly. Richard Crisp has written a paper about this with examples and he is a leading exponent of this technology. Apogee also have RBI control. I don't think any of the others do. Again I am not sure how much difference you would see in your images from this effect. I should try it out to see for myself if it is worth the trouble.

Another aspect of course is weight. Lighter is better. FLI Microline is light, Apogee Alta is medium to heavy, Apogee Ascent would be light, Starlight Express would be light, SBIG 8300 light, SBIG STL medium.

As far as QHY and Atik goes I know nothing so I can't comment.
They are up and coming makers. I had a little Atik camera once which was a cheap ToUcam type camera so they have come a long way.
There is another new maker called Morovian. Again, I know nothing about them. But the above and preceding principles still apply.

Greg.
Reply With Quote
  #12  
Old 16-05-2011, 02:21 PM
higginsdj's Avatar
higginsdj
A Lazy Astronomer

higginsdj is offline
 
Join Date: Mar 2006
Location: Canberra
Posts: 614
I've seen posts from people stating that RBI effectively makes kaf chips useless for science images. I've been using kaf chips for a decade for 1% photometry and it's not affected me as yet.

Here's a useful analysis:

http://canburytech.net/QSI532/RBI.html

The effect - at worst - seems to be measured in millimagnitudes. Not sure how many amateurs are delving into that level of precision in photometry!
Reply With Quote
  #13  
Old 16-05-2011, 05:55 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 18,183
Quote:
Originally Posted by higginsdj View Post
I've seen posts from people stating that RBI effectively makes kaf chips useless for science images. I've been using kaf chips for a decade for 1% photometry and it's not affected me as yet.

Here's a useful analysis:

http://canburytech.net/QSI532/RBI.html

The effect - at worst - seems to be measured in millimagnitudes. Not sure how many amateurs are delving into that level of precision in photometry!

I believe the KAF09000 is the worst for this.

Greg.
Reply With Quote
Reply

Bookmarks


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 10:30 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement