#1  
Old 27-04-2015, 04:42 AM
iceman's Avatar
iceman (Mike)
Sir Post a Lot!

iceman is offline
 
Join Date: Sep 2004
Location: Gosford, NSW, Australia
Posts: 36,760
Review: QSI 690i

Slawomir Lipinski has been kind enough to write a review of the QSI 690i astro imaging camera.

You can read the review on the IceInSpace Reviews page, or directly by clicking on the link below:

Review: QSI690i

Thanks very much to Slawomir for taking the time to write the review.

Last edited by iceman; 28-04-2015 at 04:37 AM.
Reply With Quote
  #2  
Old 28-04-2015, 04:37 AM
iceman's Avatar
iceman (Mike)
Sir Post a Lot!

iceman is offline
 
Join Date: Sep 2004
Location: Gosford, NSW, Australia
Posts: 36,760
Review posted.
Reply With Quote
  #3  
Old 29-04-2015, 10:43 AM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Good review Slawomir.

I am surprised the well depth is 23K at low gain. Can you change the gain on these QSI cameras in software?

The read noise is really low, that's great.

If you are not subtracting darks wouldn't it be better to simply use the cooler at 100%?

The theory of not using 100% cooling I thought was to keep the temperature regulated to match darks.

Its lucky I suppose that the camera is clean early as -10C is kind of weak cooling for a CCD. I guess I am used to FLI where -35C becomes usual but then the Starlight Expres Trius 694 cools to 45C below ambient (slightly optimistic in my use, more like 40-42C reliably).

Can you say more about the filter wheel and OAG? Is it hard to focus the OAG?

One issue I have with cameras in general is I get tired of connectivity issues when you have multiple components. It started in my SBIG days where the camera would take 10 seconds to connect as it downloaded the driver every time you started it up. I take it QSI is the same as FLI, Apogee and SX in this regard that the driver is in the on board memory of the camera? I like instant connection as I used to hate waiting 10 seconds and then it often fail and then chase down why. 10 seconds sounds stupid but its these little lags that can be irritating operating electronics.

Having a combined filter wheel and OAG in one package that connects as a unit with one click is very appealing. Its just 2 things that are no longer likely to cause you a problem (filter wheel and OAG).

I wish FLI integrated a filter wheel and OAG. Its the idea started with the SBIG STL that was so useful and convenient.

I notice QSI images are often very good. Its hard to pick one over the other with the QSI 683 Wsg 8 being so popular. I think the usual reason for chosing that over the 690 or 660 is the wider field of view.

QSI though have had some quality issues. Over at the QSI Yahoo Group there have been multiple posts about filter wheel going out of alignment, or amp glows on the 683 or artifacts. Is Kevin still running the show there or is he out? It seemed like he was the guy and if he isn't there anymore can the new guys do as good a job as he does?

Greg.
Reply With Quote
  #4  
Old 29-04-2015, 10:53 AM
lazjen's Avatar
lazjen (Chris)
PI cult member

lazjen is offline
 
Join Date: Dec 2012
Location: Flaxton, Qld
Posts: 2,064
Quote:
Originally Posted by gregbradley View Post
Its lucky I suppose that the camera is clean early as -10C is kind of weak cooling for a CCD. I guess I am used to FLI where -35C becomes usual but then the Starlight Expres Trius 694 cools to 45C below ambient (slightly optimistic in my use, more like 40-42C reliably).
If I'm reading the article correctly, he's getting the temps down to -10C from 25C+, so at at least a 35C below ambient temp. That doesn't sound too bad?
Reply With Quote
  #5  
Old 29-04-2015, 03:14 PM
Amaranthus's Avatar
Amaranthus (Barry)
Thylacinus stargazoculus

Amaranthus is offline
 
Join Date: Jan 2014
Location: Judbury, Tasmania
Posts: 1,203
I find I get marginal additional cooling on my QHY22 with 100% vs 75% cooling. Maybe 5C lower if it is a warm night, and 2C lower on a cold night. So I typically run my cooler at 75% as an optimum. On 5C nights here in Tassie, I typically cool the QHY22 to between -35C and -40C. Dark current is then virtually non-existent.
Reply With Quote
  #6  
Old 29-04-2015, 05:07 PM
Slawomir's Avatar
Slawomir (Suavi)
Registered User

Slawomir is offline
 
Join Date: Sep 2014
Location: North Queensland
Posts: 3,240
Thank you very much gentlemen for your feedback, replies and questions. I will do my best to answer questions as accurately as possible.
Quote:
Originally Posted by Amaranthus
I find I get marginal additional cooling on my QHY22 with 100% vs 75% cooling. Maybe 5C lower if it is a warm night, and 2C lower on a cold night. So I typically run my cooler at 75% as an optimum. On 5C nights here in Tassie, I typically cool the QHY22 to between -35C and -40C. Dark current is then virtually non-existent.

The cooler runs at 65-67% to cool the camera about 35C below ambient temperature(25C to -10C). I got it down to -20C (about 45C below ambient) and the cooler from memory was at about 80%. Yet there was literally no difference in the dark current between -10C and -20C, thus I run the camera at -10C and let the cooler and fans work at ease (below 70%).


Quote:
Originally Posted by lazjen
If I'm reading the article correctly, he's getting the temps down to -10C from 25C+, so at at least a 35C below ambient temp. That doesn't sound too bad?

Yes Chris, you are correct. 35C below ambient with the cooler running at 65-67% of power does not sound too bad.
Quote:
Originally Posted by gregbradley
Good review Slawomir.

I am surprised the well depth is 23K at low gain. Can you change the gain on these QSI cameras in software?


Yes, with on click of a button in Nebulosity. QSI recommends low gain when binning at 2x2 or higher, since then you physically increase the surface area of each (combined) pixel and thus space for holding electrons.
Quote:
Originally Posted by gregbradley
The read noise is really low, that's great.

If you are not subtracting darks wouldn't it be better to simply use the cooler at 100%?

The theory of not using 100% cooling I thought was to keep the temperature regulated to match darks.


There is virtually no difference in dark current between the camera running at -10C and -20C, thus I prefer the fans and cooler to run at ease with less than 70% of its capacity. Maybe I am naïve, but I like to think that equipment usually lasts longer when not running all the time at 100%...
Quote:
Originally Posted by gregbradley
Its lucky I suppose that the camera is clean early as -10C is kind of weak cooling for a CCD. I guess I am used to FLI where -35C becomes usual but then the Starlight Expres Trius 694 cools to 45C below ambient (slightly optimistic in my use, more like 40-42C reliably).

25C to -10C gives 35C difference with the cooler running at less than 70%. To achieve 45C below ambient (from 25C to -20C) the cooler was running from memory at about 80%. - I will test how low I can get with the cooler running at 100% and will place the result in here.
Quote:
Originally Posted by gregbradley
Can you say more about the filter wheel and OAG? Is it hard to focus the OAG?

One issue I have with cameras in general is I get tired of connectivity issues when you have multiple components. It started in my SBIG days where the camera would take 10 seconds to connect as it downloaded the driver every time you started it up. I take it QSI is the same as FLI, Apogee and SX in this regard that the driver is in the on board memory of the camera? I like instant connection as I used to hate waiting 10 seconds and then it often fail and then chase down why. 10 seconds sounds stupid but its these little lags that can be irritating operating electronics.

Having a combined filter wheel and OAG in one package that connects as a unit with one click is very appealing. Its just 2 things that are no longer likely to cause you a problem (filter wheel and OAG).

I wish FLI integrated a filter wheel and OAG. Its the idea started with the SBIG STL that was so useful and convenient.)


The camera connects right away - I use Nebulosity and choose QSI 500/600 from the camera menu.
I am using an external filter wheel (Orion Nautilus) and TS OAG and they are connect with T-threads. Honestly, now I wish I have gotten the WSG version instead of the slim one as it is a really nice package - but there is always room in my house for another camera in the future... :-)

Quote:
Originally Posted by gregbradley
I notice QSI images are often very good. Its hard to pick one over the other with the QSI 683 Wsg 8 being so popular. I think the usual reason for chosing that over the 690 or 660 is the wider field of view.

QSI though have had some quality issues. Over at the QSI Yahoo Group there have been multiple posts about filter wheel going out of alignment, or amp glows on the 683 or artifacts. Is Kevin still running the show there or is he out? It seemed like he was the guy and if he isn't there anymore can the new guys do as good a job as he does?

Greg.


I really cannot comment on the issues. All I can say is that QSI 690 was a substantial step up for me from my previous camera - in particular in terms of powerful cooling and build quality. The overall impression is that the guys at QSI really took time to design this camera.

Yes, there is very minor amp glow that I noticed in dark frames after heavily stretching them - but I really tried hard to see this amp glow in my light subs and even have measured them and could not see any effects of amp glow. I have seen darks and bias frames form other cameras that implemented 8300 chip and they did not look nearly as clean and even as the ones from Sony's sensor.

I hope my replies have helped to clarify a few things

Muchos Nachos
Slawomir

Last edited by Slawomir; 29-04-2015 at 06:41 PM.
Reply With Quote
  #7  
Old 29-04-2015, 06:16 PM
Slawomir's Avatar
Slawomir (Suavi)
Registered User

Slawomir is offline
 
Join Date: Sep 2014
Location: North Queensland
Posts: 3,240
Just a quick update regarding sensor cooling.

Testing has been done indoors with ambient temperature of 23.2 C and cooling set at 100%.

After 1 minute CCD temperature dropped to -4 C
After 2 minutes CCD temperature dropped to -17 C
After 3 minutes CCD temperature dropped to -22 C
After 4 minutes CCD temperature dropped to -23 C
After 5 minutes CCD temperature dropped to -23.3 C and stayed there.

That gives 23.2 +23.3 = 46.5 C below ambient at 100% cooling (indoors)
With CCD temperature set to -20 C (43.2 C below ambient) it required exactly 80% of power to maintain that temperature.
With CCD temperature set to -15 C (38.2 C below ambient) it required 69% of power to maintain that temperature.
With CCD temperature set to -10 C (33.2 C below ambient) it required 58% of power to maintain that temperature.
Reply With Quote
  #8  
Old 29-04-2015, 09:17 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Ah that sounds better. If you are not using darks then there is no need to have the temperature so stable (perhaps for bias but I doubt it changes too much).

I'd whack it up to max. -20C is pretty decent and in winter that could easily be -30 to -30C. I wonder if the read noise drops also with the lower sensor temps.

I get amp glow with the Trius as well. Its very minor. I think a lot of the guys on the QSI Yahoo site are new to CCD and they probably panic about things like that.

Good idea about a bias subtract when not using darks. I should try that.

I did use bias and flats with the AP Honders on the weekend though with the Trius 694. F3.8, 12 inch aperture and 77% QE with a super high 66% in Ha is gaining signal super fast. Mike has had easy for the last year or so with his AG12. It a good formula.

I think the smaller pixels of the QSI 690/ Trius 814 (9.2mp, 3.69 micron pixels) would work well with the Honders gives .66 arc seconds which by my reckoning is textbook ideal sampling. Especially with an AO unit.

The new QSI 6120 sounds interesting with 12mp and 3.1 arc seconds and 3 electron read noise (smaller pixels give lower read noise).

That would give .55 arc secs with the Honders. But with the small wells and the powerful 12 inch F3.8 light gathering I would be worried its too much of an experiment and it may give bloated overexposed stars. Not sure. Hard to find sample images so I'll give that a while to prove itself.

Its not encouraging to read the QSI only goes to 42,000 ADU peak instead of 65535 but then most of those levels are wasted anyway in processing.

Greg.
Reply With Quote
  #9  
Old 29-04-2015, 09:21 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
excellent review - plenty of really useful information. At 1 dark electron per 10 minutes, the cooling surely has to be good enough.

my sxH694 shows some very slight warm areas in the bias which I had put down to slight variations in the Peltier/chip-coupling. Surely the amp wouldn't need to be active while integrating? - the main amp may not even be on-chip.
Reply With Quote
  #10  
Old 29-04-2015, 09:29 PM
Slawomir's Avatar
Slawomir (Suavi)
Registered User

Slawomir is offline
 
Join Date: Sep 2014
Location: North Queensland
Posts: 3,240
Hi Greg,

Your new telescope sounds like a lot of fun, and being so fast it makes imaging much more...dynamic

Yes, max ADU values are somehow lower with QSI cameras that have Sony sensors. Mine goes up to 64000, while QSI6120 goes to 49000 - that's at least what I have read on the forums.

Thank you Ray - and you might be right regarding uneven cooling. Whatever the reason, there is a very slight gradient in my bias frames, but 10-minute darks show different pattern - there are three brighter areas near the edges of the CCD. These do not show up with shorter subs. Reinstating, these variations in ADU values for bias and dark frames are in my opinion negligible and do not affect images at all.
Reply With Quote
  #11  
Old 30-04-2015, 05:44 AM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Quote:
Originally Posted by Shiraz View Post
excellent review - plenty of really useful information. At 1 dark electron per 10 minutes, the cooling surely has to be good enough.

my sxH694 shows some very slight warm areas in the bias which I had put down to slight variations in the Peltier/chip-coupling. Surely the amp wouldn't need to be active while integrating? - the main amp may not even be on-chip.

Per Richard Crisp high cooling has other benefits beyond straight dark current suppression. Those lines you see in some Kodak full frame sensors tend to disappear or weaken heavily with high cooling. I haven't seen any artefacts in my Trius though.

Slawomir, the highest ADU is probably nothing to worry about. As we know the bulk of our image is only a small section of those 65535 levels anyway. I know QSI try to maximise the camera for each type of sensor. If you get lower noise that way then that is the way to go.

How easy is the OAG to use? Any issues or does it work fine? Is the mirrored prism large enough to get bright enough guide stars or is it a struggle? How about the filter wheel. Does it line up the same every time so flats are replicatable?

Are the cooling fans quiet and smooth or do the make a lot of noise or vibration?

The Trius fan you can't really hear and its smooth. It takes about 2-3 minutes to cool to temp which is a great feature of these small chips, they cool fast. The SX OAG is a bit of too lightweight engineered product and the stalk the guide camera attaches to can move around a little in its opening so the tightening screw has to be quite tight. It seemed to me to be sloppy engineering with too much play for a part that needs to be rigid with no motion.

Greg.
Reply With Quote
  #12  
Old 30-04-2015, 07:36 AM
Slawomir's Avatar
Slawomir (Suavi)
Registered User

Slawomir is offline
 
Join Date: Sep 2014
Location: North Queensland
Posts: 3,240
Hi Greg,

I agree that in case of QSI690 losing 65535-64000=1535 ADU is nothing to really speak of. In particular that the sensor becomes non-linear as pixels approach saturation.

My censor does not have any lines, and from my experience bias frames are independent of CCD temperature below zero Celsius, not sure whether hot/cold pixels are affected by sensor's temperature below zero Celsius neither.

Have no idea how easy it is to use the OAG - my model does not have it. I believe Rick has experience is using 683 WSG model that has an OAG integrated into the camera.

Fans are very quiet and smooth, can hardly hear them, in particular when the cooler is working at or below 80% of power.
Reply With Quote
  #13  
Old 01-05-2015, 12:21 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Bias independent of temp?
I have always matched temp for bias
I'd like to hear opinions on that
Greg
Reply With Quote
  #14  
Old 01-05-2015, 12:32 PM
Amaranthus's Avatar
Amaranthus (Barry)
Thylacinus stargazoculus

Amaranthus is offline
 
Join Date: Jan 2014
Location: Judbury, Tasmania
Posts: 1,203
I have seen a noticeable effect of temperature on bias, especially if the sensor is at ambient. I always take my bias library at the same temperature as my lights (or close to it) - once you get below about -20C, it's hard to see any additional impact on biases but vs 15C (say) it's really noticeable.
Reply With Quote
  #15  
Old 01-05-2015, 03:15 PM
Slawomir's Avatar
Slawomir (Suavi)
Registered User

Slawomir is offline
 
Join Date: Sep 2014
Location: North Queensland
Posts: 3,240
Quote:
Originally Posted by gregbradley View Post
Bias independent of temp?
I have always matched temp for bias
I'd like to hear opinions on that
Greg
Hi Greg,

I wrote "from my experience bias frames are independent of CCD temperature below zero Celsius", and this holds true (within reason) for my sensor. This might be different for Kodak sensors.

Also, I was referring to needless supercooling of Sony's sensors - past certain point extra cooling only consumes more power without really improving quality of images. Again, my opinion is based on limited experience with only two different Sony sensors. But it seems to be in line what the guys from QSI and Atik have been saying.
Reply With Quote
  #16  
Old 01-05-2015, 04:02 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
I guess it would have to be measured as there may be differences even if slight. If you have done testing I am happy to accept that.

I also wonder if read noise falls with higher cooling. I see that -110C Kingfisher V camera that never made production claimed read noise of 1.5 electrons using the Sony ICX694 chip which gets rated at 5 electrons read noise by other manufacturers.

So yes the gain from high cooling may be minimal for the Sony Interline CCDs. Perhaps not so much if Sony made full frame CCDs when RBI (ghost images) can be a factor. RBI requires an infrared preflush of several iterations and then there is a leakage of that during the exposure which adds noise. High cooling reduces that leakage and resulting noise if you use RBI preflush (I don't but some recommend it). This is full frame sensors only not interline which read the sensor differently.

At the FLI Yahoo Group there is often posts about the importance of high cooling and I have enjoyed the benefits of it with Kodak chips but yes I take your point about it being less important with Sony sensors but again I wonder about the effect of cooling on read noise, pattern noise, bias.

Some camera makers have bias drift hence the PixInsight super bias feature. I am not sure where a QSI stands on this but I do recall Rick using that feature on one of his cameras - I think it was the Apogee 16803.

I wonder if bias drift is worse with warmer temps as well. That may be harder to detect.

Higher cooling also reduces hot pixels.

Little things perhaps, but when go for optimum image quality they all add up.

Greg.
Reply With Quote
  #17  
Old 01-05-2015, 05:16 PM
Slawomir's Avatar
Slawomir (Suavi)
Registered User

Slawomir is offline
 
Join Date: Sep 2014
Location: North Queensland
Posts: 3,240
Hi Greg.

Well...in short, you were right :-)

Although improvements do diminish as you go lower in temperature, nevertheless my findings illustrate that there are extra benefits of going cooler.

I took samples of 5 bias frames at various temperatures (in 5C increments), stacked them and measured in ImageJ.

As the Mean ADU values for bias frames go down with lower temperatures, so does even more importantly the standard deviation (a measure of noise). It seems that improvements are still significant down to -10 degrees Celsius, and from then on you get very small gains at higher cooling.

Attached are simple graphs that hopefully will help to illustrate that.

The first graph just shows Mean ADU vs Temperature.

The second graph shows ADU percentage difference when going down in 5 C increments.
[For 20C data point the difference is only 3.3% because it is measured between no cooling (ambient) and CCD set cooling to 20 C. ]

The third one shows StDev vs Temperature


One can at least measure his camera and thus correct his thinking when rain has been pouring for the second day in a row...
Attached Thumbnails
Click for full-size image (Mean Bias.jpg)
57.1 KB73 views
Click for full-size image (Mean Bias percentage.jpg)
60.3 KB72 views
Click for full-size image (Mean Bias St Dev.jpg)
56.7 KB80 views
Reply With Quote
  #18  
Old 01-05-2015, 06:05 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 Slawomir View Post
Have no idea how easy it is to use the OAG - my model does not have it. I believe Rick has experience is using 683 WSG model that has an OAG integrated into the camera.
The QSI-683 that I use is in the USA (just moved from California to New Mexico) and I've never physically seen it.

We do sometimes get some odd results on bright stars with big spikes in a variety of directions. I have no idea what causes these but I have been told by one of the chaps in the group that he think it's a "feature" of the camera. I'm pretty good at removing them in PixInsight now

Quote:
Originally Posted by gregbradley View Post
Some camera makers have bias drift hence the PixInsight super bias feature. I am not sure where a QSI stands on this but I do recall Rick using that feature on one of his cameras - I think it was the Apogee 16803.
My Apogee U16M shows significant bias drift as does my SX H-18. The QSI-683 that I have been using only shows a small amount of drift. My guess is that this comes from less than perfect temperature compensation in the support circuitry for the sensor (unlike the sensor, this isn't maintained at a constant temperature.) The feature I use to counter this is overscan calibration. Super bias is a process for creating a low noise master bias from a small number of bias frames. I always take lots of bias frames so I've never used it.

Cheers,
Rick.
Reply With Quote
  #19  
Old 01-05-2015, 07:32 PM
gregbradley's Avatar
gregbradley
Registered User

gregbradley is offline
 
Join Date: Feb 2006
Location: Sydney
Posts: 17,877
Quote:
Originally Posted by Slawomir View Post
Hi Greg.

Well...in short, you were right :-)

Although improvements do diminish as you go lower in temperature, nevertheless my findings illustrate that there are extra benefits of going cooler.

I took samples of 5 bias frames at various temperatures (in 5C increments), stacked them and measured in ImageJ.

As the Mean ADU values for bias frames go down with lower temperatures, so does even more importantly the standard deviation (a measure of noise). It seems that improvements are still significant down to -10 degrees Celsius, and from then on you get very small gains at higher cooling.

Attached are simple graphs that hopefully will help to illustrate that.

The first graph just shows Mean ADU vs Temperature.

The second graph shows ADU percentage difference when going down in 5 C increments.
[For 20C data point the difference is only 3.3% because it is measured between no cooling (ambient) and CCD set cooling to 20 C. ]

The third one shows StDev vs Temperature


One can at least measure his camera and thus correct his thinking when rain has been pouring for the second day in a row...
This is a very interesting post. Thanks for testing like that. I can see what you were saying that once you hit 0C there isn't much significant change.

Usually Kodak chips list their dark current doubling with every 6 degrees C increase. Same in reverse. The bias must have some temperature sensitive component. I wonder if the main component of the bias is the read noise? That would explain the really low read noise of the Kingfisher at -110C or is it more they use a better analogue to digital converter?

Sony CMOS DSLR Exmor sensors are famous for shadow recovery and thus superior dynamic range. I wonder if some of that know how is present in the CCDs. In the CMOS chips the gain comes from every pixel having its own A/D converter as opposed to Canon which is off sensor. CCD don't have on sensor circuitry like CMOS do. But the Sony ADC may be superior to the Kodak one. Its quite likely.

Or perhaps the Sony plant runs at a much smaller resolution size on their chips like their CMOS plant does (Canon 300nm or larger, Sony something like 15!).

Greg.
Reply With Quote
  #20  
Old 02-05-2015, 09:17 AM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
interesting thread.

FWIW, the bias on my 694 also varies with temperature - not massively, but enough to matter - the main effect is an increase in the gradient. Read noise seems to be almost independent of temperature (I checked at 0, -5 and -15).

Two measurements of bias at -15 and separated by a year gave essentially the same result, apart from a few additional warm pixels (this chip seems to add 1-2 extra noticeable warm pixels a month).

Last edited by Shiraz; 02-05-2015 at 10:33 AM.
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 05:44 AM.

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