CCD ice sublimation on cold cameras.

[focaldepth]

CCD ice sublimation on cold cameras.

Is this a known phenomenon in Astro Photography or am I making it all up?

I've had my SX H18 (shutter behind the window) for about 5 years now and remember first time I used it in anger, there were circular patterns and ripples on it that slowly disappeared with each image taken. This has happened several times over the years. I originally thought it was light leakage or similar. Some times they appear on a few images in the middle of a run.

Last night after a run on M31, I turned off the cooling before bed.

This morning I turned on the cooling to take some flats and the curves and ripples came back as the camera dropped to negative temperatures. Turn off the cooling and away they went. On again and back they came....

So I left the camera busy without cooling to “dry” it out for an hour or 2. Turning on the cooling brought back the ice.

Then remembering the first time, I left the cooling on and sure enough the “ice” slowly disappeared.

A Google did not find much on ice in ccds but ice sublimation is a well known effect.

So I now think I understand what is going on and and wanted to hear if this is a common occurrence or not and am I witnessing the rapid sublimation of the ice as the temperature drops below -10C?

This hobby has so many thing to learn!

[pipnina]

Surely cooling below ambient temperatures is a bad thing for electronics since it attracts condensation? I would imagine putting a peice of electronics below freezing would make all hell break loose since the water vapour would condense onto the sensor and them freeze! (meaning it will melt once you turn it off)

I had thought the use of cooling on CCDs was to keep them at a steady temperature so you can use dark files effectively and not to take them down to crazy temperatures. But I speak from experiance only with computers and not cameras.

    ~pip

[Davey-T]

Not familiar with your camera, has it got a dessicant chamber ? may need recharging, some cameras like my Atik 4000 have a heater to stop the window misting / freezing so it can  obviously  be a problem.

Dave

[Zakalwe]

Yes, it can happen, and in a number of ways.

If the sensor chamber gets moisture into it (which can happen over time), then ice can form of the surface of the sensor. Once you start to take images, the sensor becomes slightly warmer than the cold-finger from the peltier. The ice can then sublimate off the sensor and re-form on the cold finger. I had it happen with my old QHY8L, so I developed a workflow when cooling the camera. I'd cool it in steps, over a couple of minutes, all the time driving the sensor hard by taking continuous bias type frames. This kept the sensor warmer than the cold-finger, so any moisture present would form on the cold-finger, not the sensor.

Making a perfectly sealed chamber is hard, especially when there are such large temperature changes involved. The Atiks have a dessicant plug with a long lasting dessicant to keep the chamber dry. The QHY's used an external tube that you fill with dessicant (or, you can do what I used to do and keep the camera in a sealed plastic lunchbox with dessicant beads). Some cameras purge their chambers with gasses such as argon, as these gasses will not hold moisture.

Another thing that can happen is that the outside of the sensor chamber window can become cold enough to freeze. The larger QHY cameras use a heater to prevent this from happening. Or, you can wrap a dew heater around the connecting tube to warm the air inside it slightly.

[focaldepth]

Thanks all,

Pipnina, my understanding is that the colder the better as this lowers thermal noise. I have always run at -20 deg C. Just a number I plucked out of the air.

Davet-T, no, this camera does not have a dessicant chamber.

Zakalwe, Some intering points, heating the filter wheel and coma corrector is worth a try. Hot chip, cold finger sounds good. I have my camera mounted all the time, except for a rare visual session, so it gets exposed to all seasons and humidities.

The good news is, as I say, if I keep the cooler on hard for a while the ice does eventually clear. Magic or Physics I don't know.

Just trying to undrstand what is going on.

[Zakalwe]

The good news is, as I say, if I keep the cooler on hard for a while the ice does eventually clear. Magic or Physics I don't know.

Just trying to undrstand what is going on.

The sensor generates heat as the image is captured and read off it, so it is always going to be warmer than the cold-finger. The ice will sublimate off and recondense on the finger. Try running continuous short exposures as it cools.

Can you replace the dessicant inside it? It's never god to have moisture in the chamber as it can corrode the fine wires connecting the sensor.

[Ibbo!]

a small dew band wrapped around the cam can help before you put the cooler on

[FraserClarke]

Yes, this also happens on our (old) ST8 camera. The cooling system pulls the CCD chip down fastest, so any moisture condenses there first. As the rest of the system gets colder, the ice sublimates off the chip and ends up on whatever is the coldest bit. It's a good sign the desiccant needs replaced/regenerated.  I do the same as Zakalwe -- cooling in steps to try and make the chip not the coldest bit.  I also leave the camera cooling running all the time (permanent observatory), only letting it warm up to -5C between observing sessions.

In professional instruments we spend a fair bit of effort making sure the chip (typically running at -230C to -120C, depending on the type) is never the coldest part of the cryostat, so that it doesn't get all the crud (present even in a very good vacuum) deposited on it by the same process.

Pipnina: you're right, but the idea is that the cold electronics and chip are kept in a sealed and dry compartment, so there shouldn't be any moisture to condense on them.

[ollypenrice]

CCD cameras do need something doing about this problem. The Atik Large Format cameras have heated chip windows (or can have if you deploy the system.) I leave them on by default and never have any difficulty. The equivalent SXVH36 which we also run doesn't have this feature and we lose a couple of nights a year because of icing.

Olly

[Singlin]

I have an Atik 383L and had to send it back  to be cleaned as I had so much frosting and ice.

I found that in my case at least cooling slowly in 5 °C  steps has avoided the problem.

This applies to warming up too.

I also put the desiccantt tablet in the oven at 200°C for 30 minutes every now and again.

[focaldepth]

Thanks all, so much to learn in this crazy hobby. I will take a few of these suggestions on board. It is a shame I don't have a desiccant in my camera.

...fair bit of effort making sure the chip (typically running at -230C to -120C,

Thanks ALL.

So warming the sensor and/or the window end of the camera while cooling slowly and not leaving the cooler on full cooling when not needed sounds like best practice.

I wonder if a small dehumidifier in the observatory would help in the long term as I am not keen on the idea of taking the camera off every time as I like it already focused and lined up with previous flats when I open the door and walk in