Can you get by without a cooled astro-camera if you shoot in cold temperatures?

[Trippelforge]

I have been looking around and various camera's and obviously there are price gaps between cooled and non-cooled. But I always wondered how effective the systems are in cold temperatures. I almost never drag my setup outside during the warmer months as not only will the mosquitoes eat me alive, but it's extremely humid. So I tend to go outside between October and March. By winter night time lows are normally in the 20s / 30s (or colder). 

So do cooled camera's make that much difference say at 35'F or lower? Is there a threshold before it's a must? 

[Stefan73]

You can look at various graphs for read noise vs temperature for cameras and cooler is better (although with a law of diminishing returns) but I'd say the biggest plus is being able to have a darks library.  You don't have to waste potentially over an hour of imaging time (to make 20 x 3 minute darks for example because you need to shoot them at the same temperature as your lights).  You know the lights were at -5C or whatever and the camera can reproduce that temperature at your convenience during daylight or cloud cover!

[scotty38]

It's not so much the actual temperature that becomes important but the consistency, being able to shoot all your frames at the exact same temperature. The argument now though is with the current crop of low noise cameras and whether cooling is needed at all but that's another story.....

[Elp]

Depends on the camera sensor. I've been using uncooled cameras more often but only keep exposures to within 60-120s. I could probably stretch to 180s at a push but any longer and the noise can be problematic to process out. The main benefit as mentioned is not wasting imaging time taking darks.

[carastro]

Astro photographers tend to cool their cameras to -15C or even -25C.  Obviously cooler temperatures will give you less problems, but a cooled camera will definitely give better results.

Carole 

[ollypenrice]

For the benefit of those too young to know what 35F means in sensible units, it's 1.6C!  (I've just been faffing around trying to find imperial sized bolts for my Meade SCT so this is a sore point!! 🤣)

Those temperatures will certainly help an uncooled camera and narrow the gap. I also think scotty38 is right in saying that the need for stable temperatures is diminished in modern CMOS cameras. 

For all that, given the cost of the complete rig, I'd go for cooled.

Olly

[Trippelforge]

14 minutes ago, scotty38 said:

It's not so much the actual temperature that becomes important but the consistency, being able to shoot all your frames at the exact same temperature. The argument now though is with the current crop of low noise cameras and whether cooling is needed at all but that's another story.....

 

So do the fans spin up and down to try and keep a constant temperature? I was trying to figure out how that works, and if there is a point when the fan isn't going to do anything. Hypothetically lets say it was -10'F, can the camera keep being cooled below that?

[Trippelforge]

2 minutes ago, ollypenrice said:

For the benefit of those too young to know what 35F means in sensible units, it's 1.6C!  (I've just been faffing around trying to find imperial sized bolts for my Meade SCT so this is a sore point!! 🤣)

Those temperatures will certainly help an uncooled camera and narrow the gap. I also think scotty38 is right in saying that the need for stable temperatures is diminished in modern CMOS cameras. 

For all that, given the cost of the complete rig, I'd go for cooled.

Olly

 

I will for sure just pay the difference and pick up a cooled version. Everyone made good points, and to be honest I don't ONLY image at 20-30... I mean -6 to -1 (lol). Currently it isn't that cold yet, I think this evening it's only dropping to around 10. 

 

[AstroMuni]

2 minutes ago, Trippelforge said:

So do the fans spin up and down to try and keep a constant temperature? I was trying to figure out how that works, and if there is a point when the fan isn't going to do anything. Hypothetically lets say it was -10'F, can the camera keep being cooled below that?

Its a combination of fans and peltier cooling. So as the sensor heats up during imaging, the cooling kicks in to maintain the temperature.

Having a lower outside temperature means that the cooling doesnt have to work so hard. Over and above the reasons mentioned above by others, one other reason folk prefer the cooled cameras is that its easier to take darks & flats. Hope this helps

[scotty38]

8 minutes ago, Trippelforge said:

 

So do the fans spin up and down to try and keep a constant temperature? I was trying to figure out how that works, and if there is a point when the fan isn't going to do anything. Hypothetically lets say it was -10'F, can the camera keep being cooled below that?

Basically yes, it's a bit more involved than just a fan but it'll have to work based on ambient temperature and they will only cool so far below ambient anyway. The more you ask it to cool below ambient the harder it'll have to work.

Obv over winter for most folk it'll not be an issue anyway.

Edited November 1, 2022 by scotty38

[Tomatobro]

They use solid state cooling elements. When a current flows through them one side heats up and the other cools down. The fans carry away the heat on the hot side. The fans run at a constant speed when on.

[Trippelforge]

20 minutes ago, scotty38 said:

Basically yes, it's a bit more involved than just a fan but it'll have to work based on ambient temperature and they will only cool so far below ambient anyway. The more you ask it to cool below ambient the harder it'll have to work.

Obv over winter for most folk it'll not be an issue anyway.

So they are rated via ambient temperature decrease capability? I never really looked if that was a spec or not. Just diving a bit deeper, how do people get away with using DSLR's that lack cooling? I believe you can cool them, but I assume most people just shoot with them as is. 

[ollypenrice]

41 minutes ago, Trippelforge said:

So they are rated via ambient temperature decrease capability? I never really looked if that was a spec or not. Just diving a bit deeper, how do people get away with using DSLR's that lack cooling? I believe you can cool them, but I assume most people just shoot with them as is. 

Yes, manufacturers quote a cooling performance in degrees below ambient.

You get away with it by taking much longer to produce an image and, sometimes, by accepting a lower standard at the end. You can also go for the brighter targets rather than the faint, dusty ones. DSLRs do thrive on fast F ratios, too, as Maurice Toet demonstrates. with his DSLR-Tak Epsilon images... but an AP mount, Tak Epsilon and 5DMk4 no longer constitutes a budget alternative!  It's a very portable one, though. https://www.mauricetoet.nl/DeepSky/

Olly

[Ouroboros]

I’d go for cooled if possible. For reasons of consistency and darks, as others have said.

I don’t know whether DSLRs run hotter than dedicated uncooled astro cameras, but I can recall being outside doing Astro photography at sub-zero temperatures whilst the DSLR sensor was reporting a toasty 17°C.  Lucky old thing.  Long exposures definitely heat the sensor and the temperature can rise a few degrees over the course of an imaging run.  Again introducing inconsistencies. 

As for darks you can build up a library at different temperatures. Or not bother and just rely on dithering between subs and take as many as possible during a session to take advantage of averaging.   That’s what I plumped for in the end. 

[Trippelforge]

4 minutes ago, Ouroboros said:

I’d go for cooled if possible. For reasons of consistency and darks, as others have said.

I don’t know whether DSLRs run hotter than dedicated uncooled astro cameras, but I can recall being outside doing Astro photography at sub-zero temperatures whilst the DSLR sensor was reporting a toasty 17°C.  Lucky old thing.  Long exposures definitely heat the sensor and the temperature can rise a few degrees over the course of an imaging run.  Again introducing inconsistencies. 

As for darks you can build up a library at different temperatures. Or not bother and just rely on dithering between subs and take as many as possible during a session to take advantage of averaging.   That’s what I plumped for in the end. 

 

My DSLR was jumping around like crazy a few weeks ago. I hit 46'C at one point, then it settled down to around 20'C or so. I think however my battery might've been causing issues. I had an older off-brand one that started to swell not soon after, switched back to my Canon one and its more stable now. But it seems like even 20-21'C might be a bit warm? I was never sure what kind of range was acceptable. 

I didn't know what dithering was until today, I went out and read a few articles and looked at examples... I was shocked at how big of difference it makes. I am really excited to incorporate the technique tonight!

 

[Ouroboros]

Batteries do get hot. You can avoid this by powering DSRLs externally from a suitable power supply.  In requires a battery insert. If you go that route buy a decent one. I bought a cheap one through eBay which broke after a couple of sessions. 

[Trippelforge]

10 minutes ago, Ouroboros said:

Batteries do get hot. You can avoid this by powering DSRLs externally from a suitable power supply.  In requires a battery insert. If you go that route buy a decent one. I bought a cheap one through eBay which broke after a couple of sessions. 

I came across those and they looked kind of janky. I assumed people had to used them though, I guess I should pick one up instead of buying several more batteries. 

[ONIKKINEN]

Above zero in celsius is still pretty warm, even for the low thermal signal new models (533, 571, 585, maybe others). The camera will run at least 10c warmer than ambient so youre looking at a pretty warm sensor that will be difficult to calibrate.

In places that actually get winter (at least -10c) you could get away with not cooling one of the newer models but still calibration will be tricky.

Edited November 1, 2022 by ONIKKINEN

[Trippelforge]

23 minutes ago, ONIKKINEN said:

Above zero in celsius is still pretty warm, even for the low thermal signal new models (533, 571, 585, maybe others). The camera will run at least 10c warmer than ambient so youre looking at a pretty warm sensor that will be difficult to calibrate.

In places that actually get winter (at least -10c) you could get away with not cooling one of the newer models but still calibration will be tricky.

I didn't know there was a calibration process outside of focusing like a DSLR. That makes more sense now as to why it has to be dead on. It will drop below -10C here, but not consistently enough to help me out. I am going to hold out a bit longer until I can afford one of the better cooled versions. Seems like trying to cut corners was a bad idea. 

[ollypenrice]

For many years, and maybe still, my favourite camera was/is the very noisy and low QE Atik 11000. Why did I/do I still love it? Because it produces gorgeous data which you can't kill in processing. And do I use darks? Nope. Dither? Nope. Just defect map and bias-as-dark.  Darks are the most over-rated item in the history of digital imaging, in my view.

Olly

Edited November 1, 2022 by ollypenrice

[david_taurus83]

52 minutes ago, ollypenrice said:

Darks are the most over-rated item in the history of digital imaging, in my view.

Olly

Essential though if you have the previous generation CMOS like 183, 294 and 1600 range of sensors.

I don't use darks on my 6D though it's probably the best Canon DSLR for AP. Arguably don't need darks either for the new Sony IMX sensors. 

[ollypenrice]

7 minutes ago, david_taurus83 said:

Essential though if you have the previous generation CMOS like 183, 294 and 1600 range of sensors.

I don't use darks on my 6D though it's probably the best Canon DSLR for AP. Arguably don't need darks either for the new Sony IMX sensors. 

This is to do with amp glow, I guess? In which case, yes, for sure.

Olly

[tomato]

Yes, you could light a street up with the amp glow from the ASI178, but it all calibrates out with a dark frame.

[gorann]

With the previous generation of cooled astrocameras, like the ASI1600, you definitively needed a master dark to get rid of the amp glow, but with the newer ASI2600 and ASI6200 without amp glow  I do not use darks and see no reason to use them (and I would not be surprised if they actually introduced noise).

[Ouroboros]

1 hour ago, gorann said:

With the previous generation of cooled astrocameras, like the ASI1600, you definitively needed a master dark to get rid of the amp glow, but with the newer ASI2600 and ASI6200 without amp glow  I do not use darks and see no reason to use them (and I would not be surprised if they actually introduced noise).

Do you subtract a master bias or something equivalent from the light subs?