Imaging Temperature

[Starwiz]

I've been running my ASI1600mm-Pro at -10C as it'll work all year round in this climate, but I'm wondering how much of an improvement I could expect if I ran at a lower temperature during the cooler months.

Does anyone have an experience of results run at different temperatures?

Thanks

John

[daemon]

I have most recently been running mine at -15c when ambient temp was about +12c.

The results were pretty good with much less thermal noise when compared to my DSLR.

I keep a library of darks shot at -5, -10, -15, -20, -25 so that I'm ready for different conditions. 

Tonight is expected to be cooler so I may go for -20.

Cheers,

Daemon

[SyedT]

I used to run my ASI 1600MM-C at -20 as per the graph below, taken from https://astronomy-imaging-camera.com/product/asi1600mm-cool

[Starwiz]

3 hours ago, daemon said:

I have most recently been running mine at -15c when ambient temp was about +12c.

The results were pretty good with much less thermal noise when compared to my DSLR.

I keep a library of darks shot at -5, -10, -15, -20, -25 so that I'm ready for different conditions. 

Tonight is expected to be cooler so I may go for -20.

Cheers,

Daemon

 

2 hours ago, SyedT said:

I used to run my ASI 1600MM-C at -20 as per the graph below, taken from https://astronomy-imaging-camera.com/product/asi1600mm-cool

Thanks both, I think I'll give the lower temperature a go.

John

[rickwayne]

I used to go for -20 but then I listened to Dr. Glover. I know I bang on and on about his talks but it's hard to argue with his conclusions.

YouTube video of Glover's talk

[Starwiz]

On 14/04/2020 at 04:25, rickwayne said:

I used to go for -20 but then I listened to Dr. Glover. I know I bang on and on about his talks but it's hard to argue with his conclusions.

YouTube video of Glover's talk

Thanks.  Just watched it, very informative.

John

[Knight of Clear Skies]

On 14/04/2020 at 03:25, rickwayne said:

I used to go for -20 but then I listened to Dr. Glover. I know I bang on and on about his talks but it's hard to argue with his conclusions.

YouTube video of Glover's talk

What temp do you run at now please?

[carastro]

I used to use -10 but now use -15 but I agree - 10 means you can keep it the same all year round.  But it is no problem to change the temperature to a different one as the weather warms up.

No scientific theory to this only that I feel running at very low temperatures is working the camera much harder, and therefore it might wear out more quickly, but just a theory in my own mind and I could be completely wrong about it.

Carole 

 

[Starwiz]

38 minutes ago, Knight of Clear Skies said:

What temp do you run at now please?

I've been running at -10C for the last year as the summer temperatures out here average in the early 30's C, but I've tried -20C the last couple of night, just to see if I can notice any discernible difference in the noise.

[Gina]

I run at -20°C or -25°C.  I've found lower temperatures definitely give better results.  I used to run at -30°C or -35°C but haven't managed that over this mild winter.  Now it will have to be -20°C or warmer as the ambient temperatures increase.

[Starwiz]

33 minutes ago, carastro said:

I used to use -10 but now use -15 but I agree - 10 means you can keep it the same all year round.  But it is no problem to change the temperature to a different one as the weather warms up.

No scientific theory to this only that I feel running at very low temperatures is working the camera much harder, and therefore it might wear out more quickly, but just a theory in my own mind and I could be completely wrong about it.

Carole 

 

Thanks Carole.

CMOS is quite happy running down to -55C without any issues, so the question is if places an extra burden on the cooling system.

The manufacturer would have carried out accelerated life testing during development that would have likely have used thermal temperature cycling to stress the device far beyond anything it would see in normal use.  They may even have carried out burn-in on production devices to weed out infant mortality, but that will be down to their chosen reliability strategy.  There will always be some failures running under whatever conditions the user chooses to run, but it will be a very low percentage, so don't think it's much to worry about.  Definitely use the cool-down / warm-up routines, such as in APT as these will place as little thermal stress as possible on the device.  Too much thermal stress can cause material fatigue within the chip that could manifest as a fault.  Always using the controlled cool-down / warm-up will help.

John

[Starwiz]

8 minutes ago, Gina said:

I run at -20°C or -25°C.  I've found lower temperatures definitely give better results.  I used to run at -30°C or -35°C but haven't managed that over this mild winter.  Now it will have to be -20°C or warmer as the ambient temperatures increase.

Thanks Gina.

In that case, I think I'll run for at -20C for a few more nights to so I can do a comparison with -10C frames I've collected over the last year.

How do you compare noise levels - is it just a case of zooming in on an individual frame with the same exposure, gain, etc. and doing a visual comparison?

John

[Gina]

Yes, just a subjective approach.  There's probably a way of measuring it but I haven't bothered.  That is just looking at individual subs - I use darks and flats to calibrate the light images.

[Starwiz]

5 minutes ago, Gina said:

Yes, just a subjective approach.  There's probably a way of measuring it but I haven't bothered.  That is just looking at individual subs - I use darks and flats to calibrate the light images.

Thanks Gina. 

John

[vlaiv]

I donćt think you'll see much difference, but that really depends on how, where and what you image.

Main thing about cooled cameras is set point cooling - not the temperature itself. With set point cooling you have chance of proper calibration because you control temperature.

We can sort of see what impact -10C vs -20C for ASI1600 can have if we examine dark current vs temperature graph posted above.

At -20C dark current is 0.0062e/s/px while at -10C dark current is about twice that so let's go with 0.012e/s/px. Usually doubling temperature is about 6C but if we look at a graph - in this semi log plot - in -20C to -10C line has smaller gradient then after, so 10C as doubling temperature is not far off.

Now let's examine single 300s exposure (that is long enough for CMOS).

At -20C dark current build up will be about 1.86e while at -10C it will be twice that, so 3.72e.,

Associated noise components will be ~1.364e and ~1.93e. Both very similar to read noise of 1.7e.

In 300s I guess one would accumulate enough of Sky signal to swamp both read noise and dark current noise. If that is the case - there won't be any difference in images in terms of SNR.

If you can't do -20C due to climate - don't worry, odds are that differences are minimal to -15C or -10C, because dark current is low to start with and we don't do very long exposures with CMOS sensors.

[Starwiz]

Thanks.

I'm not expecting much difference, but will do some trials just to satisfy myself and my own understanding. 

My sky is about 20 mags per arcsec sq, so I'm going with shortish subs that I've calculated to be long enough to swamp the read noise.

John