#1  
Old 07-06-2012, 09:00 PM
whzzz28's Avatar
whzzz28 (Nathan)
Registered User

whzzz28 is offline
 
Join Date: Dec 2011
Location: Brisbane
Posts: 341
KAF-8300 vs ICX694ALG

I am considering my choices for CCD to replace my DSLR.
The KAF-8300 is one of the more popular chipset used in a lot of the recommended cameras (QHY9, Sbig ST-8300 and QSI 538/638) but after doing some research i heard mixed reviews on the chip.
A lot of positives, but some negatives as well (such as QE).
So i wondered what the 8300's rival/next best chip was. The answer was: Not many rivals if any. Only one comes close to cost and mega pixels - the Sony ICX694ALG.

As i am still doing my research, i thought id ask some more seasoned people as to what they think of the Sony ICX694ALG. As far as i can tell it is Sony's newest CCD chip and only one manufacturer (Starlight Xpress) has released a camera based on the chip (one is in build from Atik i hear, but still 1year~ or more away).

The cost is so-so (~$3500 for the camera) but while most of the specs of the chip are better than the 8300, the 8300 does win over some areas.

Comparing the two, based on Starlight's details:

Kodak KAF-8300:
Pixels: 5.4uM, 3326 x 2504 (8 mega pixel)
QE: Max at 540nM (~56%), 48% at 470nM and 656nM
Readout noise: Less than 8 electron RMS - typically only 6.5
Full well capacity: 25,000 e- (unbinned)
Anti-blooming: overload margin greater than 100x
System gain: 0.35 electrons per ADU


Sony ICX694ALG:
pixels: 4.54uM, 2750 x 2200 (6 mega pixel)
QE: Max at 580nM (~77%), 50% roll-off at 360nM and 770nM
Readout noise: Less than 7 electrons RMS - typically only 5
Full well capacity: 20,000 e- (unbinned)
Anti-blooming: overload margin greater than 800x
System gain: 0.3 electrons per ADU

The Kodak has a larger well, which is important, but the QE of the Sony camera is much better than the Kodak.
The pixels are smaller which might rule it out for long focal lengths, and it is 2 less mega pixels (not that that is a huge concern). The Sony also appears to have better quality control on the chips as the acceptable defects list is much more stringent than the Kodak.

So how does the Sony chip look to others?
To me it looks to be much more sensitive than the Kodak chip, but the lower well capacity is concerning.

If you had a choice between the Kodak or Sony, which would you choose any why?
And secondly, how would you rate Starlight Xpress CCD's?

The CCD: http://www.sxccd.com/sxvr-h694

Cheers.
Reply With Quote
  #2  
Old 07-06-2012, 09:45 PM
Nortilus (Josh)
Registered User

Nortilus is offline
 
Join Date: Jan 2005
Location: Mackay, QLD
Posts: 455
also consider the camera that these chips are in and what are the differences in them also, ie - cost, cooling, ease of use and so forth...
Reply With Quote
  #3  
Old 07-06-2012, 10:14 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 15,316
An interesting comparison.

One point, make sure you are comparing apples with apples with QE. Sony uses a different QE measure than Kodak so 58% versus 77% may not be accurate.

5,000 full well difference won't mean much. My 16803 is 100,000 full well and that does make a difference over my 8300 25,000 full well.

What you find with full well is with fast systems bright stars will bloat and overexpose easily. So you counter with shorter exposure times. Not so much an issue with longer focal length but I still see it with my CDK17.

Unless your focal length is really quite long I would not worry too much about the smaller pixels.

Greg.
Reply With Quote
  #4  
Old 07-06-2012, 10:49 PM
RickS's Avatar
RickS (Rick)
PI cult recruiter

RickS is offline
 
Join Date: Apr 2010
Location: Brisbane
Posts: 10,582
Nathan,

I have a SX H-18 and I'm very happy with the results I've been getting from it. I'm using it with a Tak FSQ-106ED and a GSO RC10. I have had a couple of problems with the camera (early versions of the H-18 had a coating on the shutter blades that sheds dust!) but so far SX have been very helpful.

One issue with the KAF8300 is that it doesn't bin very well. The horizontal shift register isn't much deeper than a single sensor element, so if you try to squeeze binned data through it you will see short horizontal blooms on bright objects. If the Sony chip does this properly then that might be a point in its favour.

Have you had a look at the SX group on Yahoo? There has been some discussion of the new chips and cameras there. It's at http://groups.yahoo.com/group/starlightxpress.

Cheers,
Rick.
Reply With Quote
  #5  
Old 08-06-2012, 09:32 AM
whzzz28's Avatar
whzzz28 (Nathan)
Registered User

whzzz28 is offline
 
Join Date: Dec 2011
Location: Brisbane
Posts: 341
Quote:
Originally Posted by Nortilus View Post
also consider the camera that these chips are in and what are the differences in them also, ie - cost, cooling, ease of use and so forth...
Yes this is a good point. The Sony chipped SX camera i mentioned has some terrible looking cooling, only able to do -30 from ambient (single stage) which means, up here in summer it will be lucky to get to 0 on hot night. This pretty much makes the camera a dud for me.
-edit-
Had a good read of the yahoo group for SX. The dark frame was pretty good considering, so this chip is back on my list of possibilities.

The question becomes,
$3475 for the H694 vs $3433 for the H18 (with filter wheel and OAG).

I really do like that Sony chip, but for slightly less i can get an OAG and filter wheel on the 8300 (with +2MP)...

Tough choice (not considering other options, and i don't really want to spend $5000 on my first CCD).

-edit2-
H18 + FW + OAG = 59.5mm
A bit larger than the 55mm required by the field flattener. Is this going to be an issue?

Last edited by whzzz28; 08-06-2012 at 01:57 PM.
Reply With Quote
  #6  
Old 08-06-2012, 09:38 AM
technofetishism's Avatar
technofetishism (Nick)
Registered User

technofetishism is offline
 
Join Date: Oct 2011
Location: Adelaide, Australia
Posts: 194
Don't forget about the inherent noise in the chip design between the sony and the kodak chip. The sony chips are generally regarded as having a lower noise so the requirements for as heavy cooling (double stage in the H18 for the KAF-8300) aren't really there.

In the mailing list you can find some darkframes of the H694 have been posted.
Reply With Quote
  #7  
Old 08-06-2012, 01:13 PM
DavidN's Avatar
DavidN (David)
Registered User

DavidN is offline
 
Join Date: Jun 2012
Location: Gleneagle
Posts: 7
June 2012 update

Hey everyone

I am looking at getting a Skywatcher ED120 APO refractor and wanted to know your recommendations for a cooled ccd camera to go with it?
Colour or Mono?
When my bank balance allows I wanted to upgrade the scope to a Tak and use the same camera.
I would appreciated your thoughts on both the camera and the scope.

I was considering Orion Parsec 10100C Color, StarlightXpress SXVR-H964, others?

Thanks.

Last edited by DavidN; 08-06-2012 at 01:50 PM.
Reply With Quote
  #8  
Old 08-06-2012, 08:04 PM
whzzz28's Avatar
whzzz28 (Nathan)
Registered User

whzzz28 is offline
 
Join Date: Dec 2011
Location: Brisbane
Posts: 341
Well i guess i can rule out the QSI for good.
Bintel just hiked the price by $400
Almost as much as an FLI + FW.

This Moravian G2-8300 is looking interesting as well, if only it used 2" filters rather than unmounted 1.25" ers
Reply With Quote
  #9  
Old 09-06-2012, 08:34 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 15,316
One issue with the KAF8300 is that it doesn't bin very well. The horizontal shift register isn't much deeper than a single sensor element, so if you try to squeeze binned data through it you will see short horizontal blooms on bright objects. If the Sony chip does this properly then that might be a point in its favour.

Cheers,
Rick.[/QUOTE]

Interesting problem Rick. I have never seen any issues with my FLI ML8300 in that regard.

The small wells of the 8300 mean that bright stars do bloat quite badly in long exposures in fast systems. I counter this by using 5 minute subs instead of my usual 10 minutes. I have seen that. Now you have me wondering. I'll have to check some of my older images to see if the Luminance bloated or only the colour (I usually image colour 2x2 and luminance 1x1).

The FLI ML8300 would still be at the top of the camera list. SBIGs latest offerings may be as good or better when you consider autoguiding etc.

My experience with Starlight Express so far is they are more 2nd tier manufacturers - good not great. FLI is great, Apogee is pretty good, QSI has lots of fans, SBIG are now back in the mix with lots of new products.

Lots of choices.

If it were me the wish list would still read:

FLI Microline
SBIG STT
QSI 6 series with OAG or
SBIG STF
SBIG ST8300

I would not get an Orion or other brands unless money were tight and I was forced to.

You can get a 2nd hand ST8300 from Astromart "pretty cheaply".
Like $1500.

The KAF8300 is the best current chip in its class as far as I have seen. The Sony is quite unkown/new so it will be hard to get comparative data. I have yet to see a Sony chip match the KAF8300 so unless its way better than their earlier offerings you are pretty safe with chosing the KAF8300.

Greg.
Reply With Quote
  #10  
Old 10-06-2012, 01:26 PM
LightningNZ's Avatar
LightningNZ (Cam)
Registered User

LightningNZ is offline
 
Join Date: Oct 2011
Location: Canberra
Posts: 952
The main problem I read about with the 8300 and binning (yeah sorry I don't have one) was not bloating (cause it can drain some excess charge) but non-linearity because you simply don't get 4x the number of electrons at the amp when you bin 2x2.

Cheers,
Cam
Reply With Quote
  #11  
Old 11-06-2012, 08:17 AM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 15,316
Yes I only found out about that phenomenon a few months ago. Binning 2x2 does not necessarily mean 4X increase due to the way the chip's pixels registers are laid out.

Still I find 2x2 is definitely more sensitive with a KAF8300 clearly not 4X but probably about 2X. Same wit KAF16803.

Greg.
Reply With Quote
  #12  
Old 21-06-2012, 11:22 AM
DavidN's Avatar
DavidN (David)
Registered User

DavidN is offline
 
Join Date: Jun 2012
Location: Gleneagle
Posts: 7
Quote:
Originally Posted by gregbradley View Post
If it were me the wish list would still read:

FLI Microline
SBIG STT
QSI 6 series with OAG or
SBIG STF
SBIG ST8300

Greg.
Thanks Greg
Looks like I might try to find a FLI or SBIG.
Reply With Quote
  #13  
Old 03-07-2012, 12:08 AM
astrognosis (Eric)
Registered User

astrognosis is offline
 
Join Date: May 2012
Location: Essex, England
Posts: 9
I've just come across this thread which I find interesting and useful.

I have a C14 Edge and the new x0.7 F.R. I am planning to do asteroid and exoplanet photometry.

I have the the choice of a new SXVR H694 (4.54 um) or a QSI 532 with the Kodak 3200ME chip (6.8 um) from a freind.

Finding it difficult to decide. I will be working at F7.7 (f=2723mm)

with 2x2 binning the 694 gives 0.69 arcsec per pix

whereas the 532 gives 1.03 arcsec per pixel.

Using a 2.3 - 3 x FWHM of seeing of the 532 is better 2.3 - 3 arcsec
c.f 1.6 - 2 arcsec but a little over sampling here would be ok.

The QE of both is good. The dark current of the 694 is very good. Though the full well capacity of the 532 is much greater

Would appreciate any views or comments on suitability or preference that members might have.

With Thanks, Eric
Reply With Quote
  #14  
Old 03-07-2012, 07:27 AM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 15,316
I have a C14 Edge and the new x0.7 F.R. I am planning to do asteroid and exoplanet photometry.


Hi Eric,

You'd have to have very good seeing for either. Edge C14 even with a reducer is very long focal length. Both of those cameras have small pixels.

What this will mean is the stars will blur more with the seeing compared to a larger pixelled camera.

But then you may be more interested in lowest noise and highest QE.

Well depth is a less important aspect of cameras. Look at how popular the KAF8300 is and that has tiny well depth.

I use a CDK17 and I found imaging with an 8300 chipped camera with its 5.4 micron pixels to give inferior results to my 9 micon pixelled camera. It made for quite a drop in detail from the 8300 even though on paper the 2 had much the same QE. The 8300 images also looked slightly blurred.
But then that is for pretty pictures not measuring light dropoffs. For that I think you need the lowest noise which would be the Sony right?

Greg.
Reply With Quote
  #15  
Old 03-07-2012, 05:47 PM
Terry B's Avatar
Terry B
Country living & viewing

Terry B is offline
 
Join Date: Mar 2006
Location: Armidale
Posts: 2,759
Eric
For accurate photometry and high sensitivity you are better off with bigger pixels and non antiblooming. Buy a second hand SBIG ST10XME or ST8XME or the new SBIG versions of the same chips. Not as many megapixels but much better sensitivity.
Reply With Quote
  #16  
Old 03-07-2012, 07:55 PM
Peter Ward's Avatar
Peter Ward
Galaxy hitchhiking guide

Peter Ward is offline
 
Join Date: Dec 2007
Location: The Shire
Posts: 6,576
Quote:
Originally Posted by Terry B View Post
Eric
For accurate photometry and high sensitivity you are better off with bigger pixels and non antiblooming. Buy a second hand SBIG ST10XME or ST8XME or the new SBIG versions of the same chips. Not as many megapixels but much better sensitivity.
Very good advice
Reply With Quote
  #17  
Old 03-07-2012, 07:55 PM
billdan's Avatar
billdan (Bill)
Registered User

billdan is offline
 
Join Date: Mar 2012
Location: Narangba, SE QLD
Posts: 1,305
One item that hasn't been mentioned is the physical size of these chips.

A DSLR chip is about 25.1 x 17.6mm
The KAF8300 is 17.9 x 13.5mm
The ICX694 is 12.5 x 10mm

Thats a lot of light shooting past that Sony chip, it has effectivly halved your aperture, compared to your DSLR.

Regards

Bill
Reply With Quote
  #18  
Old 04-07-2012, 03:35 PM
LightningNZ's Avatar
LightningNZ (Cam)
Registered User

LightningNZ is offline
 
Join Date: Oct 2011
Location: Canberra
Posts: 952
Bill - No, it hasn't changed your aperture at all. It's only changed your FOV. That's why APS-C sized chips have a so-called "magnification factor" for any lens attached to them - the field of view is reduced by that much and spread over all your pixels.

For photometry the bucket size matters a lot as it is one of the main ways to make sure you have a wide dynamic range (and hopefully good linearity over that range). Given that the whole point of doing CCD photometry is that you get photometry over a wide field then with a big dynamic range you have more comparison stars to work with.

Hope this helps,
Cam
Reply With Quote
  #19  
Old 04-07-2012, 04:09 PM
billdan's Avatar
billdan (Bill)
Registered User

billdan is offline
 
Join Date: Mar 2012
Location: Narangba, SE QLD
Posts: 1,305
Hi Cam,
Sorry I meant to say FOV and not aperture.

I only wanted Nathan to be aware he would lose his wide FOV if he went the Sony ICX694 path.

Thanks mate
Bill
Reply With Quote
  #20  
Old 04-07-2012, 05:20 PM
whzzz28's Avatar
whzzz28 (Nathan)
Registered User

whzzz28 is offline
 
Join Date: Dec 2011
Location: Brisbane
Posts: 341
Thought i should make a few updates here;

There is a typo in my first post but i don't appear to be able to edit it anymore.
Anti-blooming on the KAF8300 is in fact 1000x not 100x. You can thank Starlight Xpress' webpage for that typo.

Also, Atik have in fact released their 694 camera - Atik 460EX.
It is compared with the SX camera over here: http://blog.astrofotky.cz/pavelpech/?page_id=782

The Atik version seems to rate very well, where as the SX version has a pretty bad set of results.

I have since expanded my views even further and whilst i liked the Sony have decided that either the KAF8300 or KAI4022 would suit me better. At this stage the KAF8300 is winning purely due to cost, but the KAI4022 is a nice looking chip.
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 08:12 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Meade Australia
Advertisement
Lunatico Astronomical
Advertisement
Celestron Australia
Advertisement
SkyWatcher Australia
Advertisement
OzScopes Authorised Dealer
Advertisement
NexDome Observatories
Advertisement
Astronomy and Electronics Centre
Advertisement