£900 to spend on a dedicated cooled camera - What to get??!!

[AndyThilo]

Hi

Decided to use my bonus to get a dedicated cooled camera. I don't have time for mono or the money for filters/wheel to go with it so only looking at colour. Considering the ASI294MC Pro, the new ASI533 although it's not out yet, ASI183MC Pro. This is go on my Skywatcher Esprit 80 and replacing my 60D. 

I'm leaning towards the 294 but should I wait for the 533?, any other options to consider? This will be a one-off purchase so I need to get it right. It's purely for DSO, although planetary videoing would be cool if it's possible with them. 

Cheers

Andy

[Dinglem]

This may be of interest " The NEW ZWO ASI 533MC-PRO camera is now available to buy at FLO 🙂 "

I've been thinking of getting a 294 but not sure now!!!

[AndyThilo]

7 minutes ago, Dinglem said:

This may be of interest " The NEW ZWO ASI 533MC-PRO camera is now available to buy at FLO 🙂 "

I've been thinking of getting a 294 but not sure now!!!

Yeah but a couple of things put me off, the square sensor makes for tricky composition on targets like Orion, and from the graphs the QE on the 294 is higher than the 533. Also not sure on delivery date as some places are stating Jan next year!

[AndyThilo]

Plus Andromeda won't fit in the frame with my Esprit 80 and a 533 and I really want to do that again.

[sloz1664]

2 hours ago, AndyThilo said:

Yeah but a couple of things put me off, the square sensor makes for tricky composition on targets like Orion, and from the graphs the QE on the 294 is higher than the 533. Also not sure on delivery date as some places are stating Jan next year!

I can't understand ZWO's logic on a square sensor. All computer / tv screens are widescreen, which lends itself to rectangular sensors.

Steve

[Alien 13]

5 minutes ago, sloz1664 said:

I can't understand ZWO's logic on a square sensor. All computer / tv screens are widescreen, which lends itself to rectangular sensors.

Steve

Ideally all sensors should be round but square is the next best thing, you are rejecting an awful lot of data with a rectangular one...

Alan

[pete_l]

14 minutes ago, sloz1664 said:

I can't understand ZWO's logic on a square sensor. All computer / tv screens are widescreen, which lends itself to rectangular sensors.

Steve

Trying to fit a rectangluar sensor into / over a circular field of view wastes an awful lot of something. Either of expensive sensor area or expensive image circle. The wider and flatter a sensor is, the worse the area of wastage.

For reasons that will not become apparent I have been looking at sensors that would be the best "fit" (square peg / round hole type fit) for a Night Owl F/4 FR. This has an image circle of 16mm, so an image area of about 200mm²
A ZWO533 is a pretty good match to this, being completely inside the circle, but that only uses 60% of the image circle.
A 4:3 sensor that was the same height as the N.O's circle would be 21x16mm and have an area of 336mm² So being 68% too big.
A 16:9 sensor with the same height is 28.5x16mm which is 456mm² so more than half the sensor area is unusable.

The most efficient shape is therefore a square. What would be even better is if telescope optics produced a square or rectangular image, too

 

[AndyThilo]

Well hopefully my ASI294MC Pro arrives tomorrow with a UV cut filter. 

[wimvb]

On 26/11/2019 at 21:57, sloz1664 said:

I can't understand ZWO's logic on a square sensor. All computer / tv screens are widescreen, which lends itself to rectangular sensors.

Steve

A proven concept, I'd think 😉

https://sv.m.wikipedia.org/wiki/Victor_Hasselblad_(företag)#/media/Fil%3AHasselblad_1600F.jpg

No need to rotate the camera and no one telling me that images must be landscape oriented. Just crop the final image to get the best composition. 

But back to the original question : the 533 promises amp glow free images. That might be worth considering. 

[ollypenrice]

On 26/11/2019 at 19:03, AndyThilo said:

Hi

Decided to use my bonus to get a dedicated cooled camera. I don't have time for mono

Your point about the cost of filters is valid but let's not perpetuate the myth that OSC is faster than mono. It is about 20% slower in basic broadband imaging and way, way slower than that in narrowband. 

If you take a deep M31 you won't fit it on any chip in regular amateur use. Using our full frame (36mm wide) Atik 11000s at 530mm focal length it still needed two panels. The deeper you go the bigger it gets!

Enjoy the new camera.

Olly

[Adam J]

I would not even have to think about it, if I had about 900 to spend it would be the ASI183mm pro, failing that and going for a OSC it would 100% be the 533, no amp glow and comes with a duo band filter (or at least I assume it will in the UK? as it will in the US). Apart from that I like the pixel pitch better and as Olly says you need two frames for M31 with any of these sensors. In that case you might consider the 533 more effective as it will provide a nice 2:1 frame ratio in two panels which prints well should you want. 

Adam

[sloz1664]

8 hours ago, wimvb said:

A proven concept, I'd think 😉

https://sv.m.wikipedia.org/wiki/Victor_Hasselblad_(företag)#/media/Fil%3AHasselblad_1600F.jpg

No need to rotate the camera and no one telling me that images must be landscape oriented. Just crop the final image to get the best composition. 

But back to the original question : the 533 promises amp glow free images. That might be worth considering. 

Imaging sensors were originally manufactured for the photographic community and they haven't used a square format since the old 2 1/4" black & white film. The majority of astro sensors are used in photographic sensors, basically to reduce cost.

 

[Adam J]

45 minutes ago, sloz1664 said:

Imaging sensors were originally manufactured for the photographic community and they haven't used a square format since the old 2 1/4" black & white film. The majority of astro sensors are used in photographic sensors, basically to reduce cost.

 

Yes and the reason this one is square is because its designed for use in 360 degree video applications. 

[DaveS]

The last small sensor square format cameras were based around the KAI4022, which had several attractions for me, but the low QE and high read noise put me off, plus they were un-cheap. The ASI533 looks very attractive, if a little small. No, I cannot afford one of the KAF16803 cameras, let alone a Kepler 4040 (Or even the 6060 )

Square format has a lot to commend it for getting the most from the image circle, plus you can ignore questions of orientation. I already have an ASI183 on my TS 80mm f/4.4 but am seriously thinking of swapping it for an ASI533. If only they did a mono version *sigh*.

BTW, you can still buy 120 film.

[Adam J]

22 minutes ago, DaveS said:

The last small sensor square format cameras were based around the KAI4022, which had several attractions for me, but the low QE and high read noise put me off, plus they were un-cheap. The ASI533 looks very attractive, if a little small. No, I cannot afford one of the KAF16803 cameras, let alone a Kepler 4040 (Or even the 6060 )

Square format has a lot to commend it for getting the most from the image circle, plus you can ignore questions of orientation. I already have an ASI183 on my TS 80mm f/4.4 but am seriously thinking of swapping it for an ASI533. If only they did a mono version *sigh*.

BTW, you can still buy 120 film.

If they make a mono version ill be swapping out my ASI1600mm pro for one. 

[AndyThilo]

On 28/11/2019 at 08:48, ollypenrice said:

Your point about the cost of filters is valid but let's not perpetuate the myth that OSC is faster than mono. It is about 20% slower in basic broadband imaging and way, way slower than that in narrowband. 

If you take a deep M31 you won't fit it on any chip in regular amateur use. Using our full frame (36mm wide) Atik 11000s at 530mm focal length it still needed two panels. The deeper you go the bigger it gets!

Enjoy the new camera.

Olly

I guess my main point was the cost. To add a filter wheel and 3 dedicated filters if not financially possible for me at the moment. Obviously in the future I'd like to go full narrowband, but as I've only just started, I need something I can grow with and is easy. I would have stuck with my 60D if it weren't for the fact I got the 294MC Pro for effectively £350. 

I'm not sure I follow with the 'fit on any chip'. The object size in the frame is related to the sensor size and focal length of the scope. 

[wimvb]

7 minutes ago, AndyThilo said:

I'm not sure I follow with the 'fit on any chip'. The object size in the frame is related to the sensor size and focal length of the scope

You also have to factor in pixelscale (arcseconds/pixel) to get "deep" detail. Obviously with the right focal length, you can fit M31 on even the smallest sensor, but it won't become a "good" image in terms of preserving faint detail, because the pixelscale is too coarse. If you want detail and still capture all of M31 (even the fainter dusty regions), you would need a large sensor with very small pixels, and enough dynamic range. You might be hard pressed to find a camera with such a sensor. Although some CMOS cameras with 2.4 um pixels come close.

[DaveS]

Even in your example above,  M31 is bleeding off the corners, and I suspect that one of Olly's really deep images would be severely cropped.

[AKB]

On 26/11/2019 at 21:20, pete_l said:

The most efficient shape is therefore a square. What would be even better is if telescope optics produced a square or rectangular image, too

Surely, it would be so much more efficient if the chip were octagonal?  Easy to make too, just knock the corners off.  A really good match to a circular FOV.

[ollypenrice]

4 minutes ago, wimvb said:

You also have to factor in pixelscale (arcseconds/pixel) to get "deep" detail. Obviously with the right focal length, you can fit M31 on even the smallest sensor, but it won't become a "good" image in terms of preserving faint detail, because the pixelscale is too coarse. If you want detail and still capture all of M31 (even the fainter dusty regions), you would need a large sensor with very small pixels, and enough dynamic range. You might be hard pressed to find a camera with such a sensor. Although some CMOS cameras with 2.4 um pixels come close.

I was being a bit careless in my wording. What I meant was that, even in smaller telescopes, the chips available will struggle to fit M31 without cropping. Also most smaller telescopes (indeed most telescopes) struggle to cover a full frame chip. The image you posted shows that both ends of the galaxy have been cropped and that's not the deepest M31 that it's possible to take. Below you can see that quite a lot is excluded by your frame simulation.

Olly

[Adam J]

The shape is not really the main advantage of this its the zero amp glow, yes amp glow on other chips calibrates out but do you really think that it has no effect on Signal to noise in the areas under that amp glow? If you look hard in images you can even see it in the centre of the right edge of the 294 as an area of increased noise, admittedly you do have to go looking for it though and its more apparent in images with lower total integration. 

Edited November 29, 2019 by Adam J

[AndyThilo]

4 minutes ago, ollypenrice said:

I was being a bit careless in my wording. What I meant was that, even in smaller telescopes, the chips available will struggle to fit M31 without cropping. Also most smaller telescopes (indeed most telescopes) struggle to cover a full frame chip. The image you posted shows that both ends of the galaxy have been cropped and that's not the deepest M31 that it's possible to take. Below you can see that quite a lot is excluded by your frame simulation.

Olly

I understand what you're saying, I guess it's why people have multiple scopes. Widefield for Andromeda/Orion, Deep for Dumbbell etc.. At the moment I have my Esprit 80, maybe in the future I can get another or sell the 80 and get a wider one and longer one. 

[vlaiv]

I would like to point out something here that might not be obvious (for me at least - it took some time to realize it) - larger sensor is faster sensor

Most of recent talk about speed of setup was in terms of pixel scale, and I'm to large extent guilty of being promoter of such view. It is in fact matter of aperture at resolution. However, most of the time that discussion neglects one important thing - resolution is not set in stone. It is for the most part if we follow traditional processing workflow - but then again, why should we, if there is alternative that allows for more flexibility?

Before I explain what I meant with above, we need to assert one observation - when doing software binning of pixels, only difference between camera with smaller pixels and one with larger pixels (everything else being equal) is amount of read noise. There is rather nice way to control impact of read noise on the final result - exposure length. Longer exposure length comes with it's challenges (like guiding, probability of wasted data, etc ...), but in principle one can control impact of read noise.

Back to the story of faster sensor. If we have larger sensor - we can use larger scope (both aperture and focal length) to cover same FOV. If we control our pixel scale - for example thru fractional binning or similar techniques, we can image - same FOV at same resolution (in terms of "/px) using larger sensor on a larger scope. Larger scope will gather more light - resulting SNR will be better with larger sensor. Larger sensor is faster (it does hover mean certain processing techniques, being aware of read noise management and pairing it up with suitable telescope for target FOV and matching resolution via suitable binning).

In that sense 294 is faster than both 183 and 533 (at a small price premium).

Just to add - I'm in general not overly concerned with amp glow in CMOS sensors - from what I saw, it calibrates out and level of noise injected is comparable to read noise level which is rather low on CMOS sensors anyway.

[AndyThilo]

Not only that, a larger size sensor with less MP in theory has less noise. 

[Michael Hogan]

ASI294mc Pro or ASI533 two best camera,s