Best astro cameras out there

[gregbradley]

I am looking at perhaps getting another camera (oh gawd not another one) for my setup.

I see Roland Christen of Astrophysics uses a QSI 683 wsg 8300 for his Honders. It does work out to .96 arc seconds per pixel which he recommends for the average user (assumes 3 arc second seeing).

I have been impressed with QSI8300 images and the built in OAG and filter wheel. Additionally although the 8300 is not a good match for my CDK17 the short backfocus of 36mm means I could use it with the reducer which has an optimum spacing of 44mm (real workable right 44mm what were they thinking of??).

I have also thought the KAI16070 looks like a nice sensor and images from the Cedic team using that on an RCOS 14.5 were very good indeed.
Whereas images using the FLI Proline 29050 mostly looked terrible. Not a comment on FLI cameras at all as they are great - just the sensor.

The QSI with the 4022 sensor with 7.4 micron pixels may be a good compromise. It is 15mm square so a bit bigger than the 8300 sensor, slightly lower QE than the 8300 but not much at 55% versus 60.
Bigger wells, probably better dynamic range. It may suit the CDK17 with reducer as well as work with the Honders at 1.32 arc sec/pixel versus .96 arc secs/pixel with the 8300.

The 694 sensor is very good but a tad small. It works well on the Honders at .81 arc secs/pixel and I think its ideal for small galaxies and nebulas with its superior read noise, high QE in narrowband (twice as sensitive to O111 than the famous 6303 sensor).

The 6303 sensor is always interesting. Blooming though sounds like a pain and I can always do subframing on my 16803 to emulate a 6303 with mostly superior performance to the 6303 (lower read noise, slightly lower QE but close and no blooming, larger wells, longer lasting sensor).

Sony makes an APS sized sensor but it one shot colour only. Bugger. It would be perfect. Maybe some brave person can debayer one.

STXL11002 with AOX is appealing. Martin Pugh loves his and its a proven performer. Reliability and mature system though is unclear with the takeover. Also I think it suits those who have mega hours available to overcome the lower QE at 50% max and 30% in Ha or less. It would work well on the CDK17.

FLI also make a Microline 16050 which is APS sized. It has slightly lower QE, more like an 11002 but APS is a good size for me I think. That is a possibility. If it doesn't work out though resale value would be a concern as its not a known type of sensor.

What do you think is a good camera at the moment?

Greg.

[TR (Terry)]

Simple, one that gets used.

Terry

[Peter.M]

Originally Posted by gregbradley

I am looking at perhaps getting another camera (oh gawd not another one) for my setup.

I see Roland Christen of Astrophysics uses a QSI 683 wsg 8300 for his Honders. It does work out to .96 arc seconds per pixel which he recommends for the average user (assumes 3 arc second seeing).

I have been impressed with QSI8300 images and the built in OAG and filter wheel. Additionally although the 8300 is not a good match for my CDK17 the short backfocus of 36mm means I could use it with the reducer which has an optimum spacing of 44mm (real workable right 44mm what were they thinking of??).

I have also thought the KAI16070 looks like a nice sensor and images from the Cedic team using that on an RCOS 14.5 were very good indeed.
Whereas images using the FLI Proline 29050 mostly looked terrible. Not a comment on FLI cameras at all as they are great - just the sensor.

The QSI with the 4022 sensor with 7.4 micron pixels may be a good compromise. It is 15mm square so a bit bigger than the 8300 sensor, slightly lower QE than the 8300 but not much at 55% versus 60.
Bigger wells, probably better dynamic range. It may suit the CDK17 with reducer as well as work with the Honders at 1.32 arc sec/pixel versus .96 arc secs/pixel with the 8300.

The 694 sensor is very good but a tad small. It works well on the Honders at .81 arc secs/pixel and I think its ideal for small galaxies and nebulas with its superior read noise, high QE in narrowband (twice as sensitive to O111 than the famous 6303 sensor).

The 6303 sensor is always interesting. Blooming though sounds like a pain and I can always do subframing on my 16803 to emulate a 6303 with mostly superior performance to the 6303 (lower read noise, slightly lower QE but close and no blooming, larger wells, longer lasting sensor).

Sony makes an APS sized sensor but it one shot colour only. Bugger. It would be perfect. Maybe some brave person can debayer one.

STXL11002 with AOX is appealing. Martin Pugh loves his and its a proven performer. Reliability and mature system though is unclear with the takeover. Also I think it suits those who have mega hours available to overcome the lower QE at 50% max and 30% in Ha or less. It would work well on the CDK17.

FLI also make a Microline 16050 which is APS sized. It has slightly lower QE, more like an 11002 but APS is a good size for me I think. That is a possibility. If it doesn't work out though resale value would be a concern as its not a known type of sensor.

What do you think is a good camera at the moment?

Greg.

Maybe you should spend less time going over spec sheets contemplating which would give you a 1% increase in this or that parameter and spend the cash getting your stuff operating remotely so you can take advantage of EVERY free night at your dark site?

Just my 2 cents, you could have a scope that collects light 3 times faster than a FSQ, with a cam that is twice as sensitive. But if you only get out there one out of every 10 nights, you are still going to get less data than someone who operates inferior gear every clear night.

[Atmos (Colin)]

I personally see "best" as being a very subjective term, it really comes down to what you're wanting to do and what sample size you want. Once you have a handle on both of those, then you can pick a camera from that narrowed down list.

Want to do large sky areas, you'll be happy with an image scale of 1.5-2.5. If however you want to resolve individual stars in cores of globular clusters, you'll want to be closer to 0.5 arcsec/pixel. Personally, I think image scale should always be the first place to start, you start with what you want your camera to do with your telescope.

As for one camera that I would go for, it would be the FLI-16803.

[Slawomir (Suavi)]

Hi Greg,

It won't be easy to decide but that is part of the fun!

If I may...camera specs are very important, in particular that you will use it with your beautiful scopes. Also, having a complete package (dedicated wheel and OAG) would also be crucial - we don't won't to have flexes and misalignments.

Whatever you choose, it is always exciting to get a new camera!

Regards
S.

[LewisM]

Use an 8300 and just mosaic ya lazy sod!

[gregbradley]

Originally Posted by Atmos

I personally see "best" as being a very subjective term, it really comes down to what you're wanting to do and what sample size you want. Once you have a handle on both of those, then you can pick a camera from that narrowed down list.

Want to do large sky areas, you'll be happy with an image scale of 1.5-2.5. If however you want to resolve individual stars in cores of globular clusters, you'll want to be closer to 0.5 arcsec/pixel. Personally, I think image scale should always be the first place to start, you start with what you want your camera to do with your telescope.

As for one camera that I would go for, it would be the FLI-16803.

Yes FLI PL16803 is awesome. That is my primary camera.
I was thinking about using the subframing ability to have it take only a smaller part of the chip in those situations where I want galaxies etc. It saves on file sizes and processing time.


Greg.

[gregbradley]

Originally Posted by Peter.M

Maybe you should spend less time going over spec sheets contemplating which would give you a 1% increase in this or that parameter and spend the cash getting your stuff operating remotely so you can take advantage of EVERY free night at your dark site?

Just my 2 cents, you could have a scope that collects light 3 times faster than a FSQ, with a cam that is twice as sensitive. But if you only get out there one out of every 10 nights, you are still going to get less data than someone who operates inferior gear every clear night.

What makes you think I don't have access to every clear night already? Where did you get that idea from?

I do. I have an observatory at home which is just outside the house. I live in semi rural skies that are reasonably dark (especially to the west), I don't need to go anywhere.
It also has 2 setups in it as its quite large so I can run 2 systems at once which is what I am setting up so I can get 2 lots of data from the same night.

I also have a totally dark site that I go to occasionally and setting that up is an appealing prospect. It would mean a solar power supply, a proper internet connection, a dome etc. A long term project perhaps. I also enjoy the hands on aspect of handling the gear and being out under the skies.

The dark site does have the advantage of a larger number of clear nights than at home. But that is not always true like at the moment the clear winter nights and drier weather seems to be giving a nice run of clear nights.

Greg.

[gregbradley]

Originally Posted by Slawomir

Hi Greg,

It won't be easy to decide but that is part of the fun!

If I may...camera specs are very important, in particular that you will use it with your beautiful scopes. Also, having a complete package (dedicated wheel and OAG) would also be crucial - we don't won't to have flexes and misalignments.

Whatever you choose, it is always exciting to get a new camera!

Regards
S.

I don't need one but I am always looking for a good combo and what works ideally. Good point about the flex. I don't know that flex has been an issue but certainly the QSI design is a smart one and I did not think of that aspect of them.

Originally Posted by LewisM

Use an 8300 and just mosaic ya lazy sod!

Hehehe. Yes the 8300 is one of the better ex-Kodak sensors.

I'll do a mosaic just for you Lewis!

Greg.

[Bassnut (Fred)]

Since you have a 16803 which is pretty much hard to beat for general purpose, why muck around with other stuff that just changes the spec a bit....for general purpose. Step out of the box and get mad different. Specialise for fun, do stuff others cant, cream something hard, produce pics way out of left field for a change.

1/ A monster chip on the Honders for pics that looks fantastic on a 4k monitor, thats different. Not my area of expertise.
2/ Since you have a dark site and endless exposure opportunity, go stupid deep different with NF NB megadata, really nail it on the CDK. Do it properly with a NABG cam or rear illuminated ultra high QE with 3nm filters. Even if it takes a month, you will present a pic NO ONE else can do. Whats better, a bunch of generic shots or a cracker occasionally to be really proud of!.

Id recommend the 6303, (but then I would!). It does have way higher QE at Ha than the 16803.

Anyway Greg, buy a cam that aids you to provide DIFFERENT pics, youve already done all the generic options.

[Peter Ward]

Originally Posted by gregbradley

STXL11002 with AOX is appealing. Martin Pugh loves his and its a proven performer. Reliability and mature system though is unclear with the takeover. .........

I am pleased to say Cyanogen have cleaned up the mess left behind by the Aplegen management.

Recent production STXL units have shown better performance and reliability.

[Shiraz (Ray)]

+1 with Fred - do something different.

with this http://www.flicamera.com/51.html and your RHA you could do 50mP seeing-limited images - that's getting into large survey scope territory. If you felt the need, you might even consider drizzle stacking to get a full 200mP and hang your prints up on the harbour bridge maybe. Just a thought...

[Bassnut (Fred)]

Originally Posted by Peter Ward

I am pleased to say Cyanogen have cleaned up the mess left behind by the Aplegen management.

Recent production STXL units have shown better performance and reliability.

Ive had my STXL6303 powered up literally 24/7 for 2 yrs now, taking pics on every clear night, never a missed beat new straight out of the box , no hint of the need for a dew soaker recharge either. The perfect no-maint remote machine IMO And, -35deg year round, easily, regardless of ambient.

[RickS (Rick)]

Originally Posted by Shiraz

with this http://www.flicamera.com/51.html and your RHA you could do 50mP seeing-limited images - that's getting into large survey scope territory. If you felt the need, you might even consider drizzle stacking to get a full 200mP and hang your prints up on the harbour bridge maybe. Just a thought...

Ray,

Looks like a great camera but the 11.5e- read noise spec is a bit scary for narrowband use.

I'm sitting on the sidelines here. I don't think we've had a significant advance in astro sensor technology for a while. If you want a big FOV then there's nothing much better than a KAF-16803. If you want high QE and low noise and can live with a small FOV then the ICX-694 is great. Apart from that the current offerings are pretty unexciting unless you can afford exotic silicon.

Cheers,
Rick.

[RickS (Rick)]

Originally Posted by Bassnut

Really!, the 6303 is 15e read noise and I just dont care. Ive had a couple, and im feeling stropy . Enough with that stuff, its not scary with NB. At no time did I think "****, I just processed this awesome NB pic with a scary 15e read noise, I need to delete it as fast as I can".

That's very devil-may-care of you Fred but when you sober up it may occur to you that mathematics trumps attitude when it comes to astrophotography

My camera has 8e- read noise. If it had 15e- read noise then I'd need to do 2 hour narrowband subs to match the results I currently get in 30 mins. That's cool if you have a permanent set up and can do 4 times the exposure time but I'm a nomadic imager with my home set up and I really care about imaging efficiency. So for me read noise sucks... I wish I could get a large sensor with only a few e- read noise like the ICX-694. Somebody could become a thousandaire from the sales to AP folks if they created a camera like that

Cheers,
Rick.

[Shiraz (Ray)]

Originally Posted by RickS

Ray,

Looks like a great camera but the 11.5e- read noise spec is a bit scary for narrowband use.

I'm sitting on the sidelines here. I don't think we've had a significant advance in astro sensor technology for a while. If you want a big FOV then there's nothing much better than a KAF-16803. If you want high QE and low noise and can live with a small FOV then the ICX-694 is great. Apart from that the current offerings are pretty unexciting unless you can afford exotic silicon.

Cheers,
Rick.

agreed that is a fair bit of noise for NB, but the speed of Greg's RHA would compensate a bit. Should be a very interesting broadband setup.

TV resolution keeps going up, so maybe Sony will eventually bring out a pro video cam chip with lots of moderately sized super low noise pixels - lets hope .

[Slawomir (Suavi)]

Hi Greg,

I am probably mentioning something obvious...but since you have a nice large aperture and fast f-ratio, it would make sense to add to that a camera with low read noise. In this way you would be able to pick even fainter bits of glowing cosmic matter. Anyway, just sharing my thoughts.

Muchos Nachos
S.

[gregbradley]

Originally Posted by Bassnut

Since you have a 16803 which is pretty much hard to beat for general purpose, why muck around with other stuff that just changes the spec a bit....for general purpose. Step out of the box and get mad different. Specialise for fun, do stuff others cant, cream something hard, produce pics way out of left field for a change.

1/ A monster chip on the Honders for pics that looks fantastic on a 4k monitor, thats different. Not my area of expertise.
2/ Since you have a dark site and endless exposure opportunity, go stupid deep different with NF NB megadata, really nail it on the CDK. Do it properly with a NABG cam or rear illuminated ultra high QE with 3nm filters. Even if it takes a month, you will present a pic NO ONE else can do. Whats better, a bunch of generic shots or a cracker occasionally to be really proud of!.

Id recommend the 6303, (but then I would!). It does have way higher QE at Ha than the 16803.

Anyway Greg, buy a cam that aids you to provide DIFFERENT pics, youve already done all the generic options.

Yes you are quite right and something to consider.

Originally Posted by Peter Ward

I am pleased to say Cyanogen have cleaned up the mess left behind by the Aplegen management.

Recent production STXL units have shown better performance and reliability.

Ah, that's great to hear. I figured that was the likely scenario. Doug George even though I don't know much about him seems like a pretty on the ball sort of guy to make his own acquisition software that becomes super popular.

Greg.

[gregbradley]

Originally Posted by Slawomir

Hi Greg,

I am probably mentioning something obvious...but since you have a nice large aperture and fast f-ratio, it would make sense to add to that a camera with low read noise. In this way you would be able to pick even fainter bits of glowing cosmic matter. Anyway, just sharing my thoughts.

Muchos Nachos
S.

Which one are you thinking of? The QSI690? It sounds appealing but its quite a small FOV which as Mike has shown is not that much of a problem with the 305 F3.8 large FOV, I have the 694 and its low read noise and highly sensitive. I think it would be good for galaxies and its not that great a match for the CDK17 though. The 6303 might be a better match for the CDK or the 4022 chip would be in between compromise for both CDK17 galaxy work and the Honders for smaller objects.

The 51mp FLI Proline is an interesting looking camera. Over 60% QE at peak is impressive. 6 micron pixels would be good for the Honders and not so good for the CDK. Perhaps OK with the reducer but that means back to a guide scope. Perhaps an APM Riccardi reducer may work on the CDK but its a gamble. An AP CCD67 didn't work that well on the CDK. It came to focus but had noticeably degraded resolution over the Planewave reducer.

Greg.

[Slawomir (Suavi)]

Hi Greg, actually I did not have any camera in mind when I suggested one with low read noise. I just thought it could be an important factor to keep in mind.

That 50mp one sounds very impressive, I think you could always bin in for the CDK? That would yield over 12mp 12 micron pixels, and perhaps well depth would about double too when binned?