How quick should camera cooldown and warmup be?

I have just bought my first astro camera (having previously used a DSLR).

APT has a cooling (and warming) aid which allows you to change the temperature in specific stages with delays in between. I am told that too rapid a change can damage the sensor.

But what is 'too rapid'?

Is it ok to just set a target temp and let the camera do its own thing for cooldown? Then would it be ok to let it just warm up naturally when disconnected?

Thanks!

Hi Stuart,

Honestly I am not aware of a too rapid change that damages the sensor. The speed is always limited by the capacity of the fan to cool the camera and the delta between ambient/target temperature. Even when you set your temp without a time objective, it will take a few minutes for it to cool. For example in SGPro I was set to something like 3 or 5 min and a target of -20°C, but if the ambient is too high, it never achieves the target in 3 minutes. So just connect it, set your objective and by the time you finish polar aligning or focusing/plate solving it should be ok.

CS

Bogdan

The dynamic behaviour of the cooling system (which I assume is made of a peltier device and a fan) is dictated by the available power and the thermal inertia of the camera/sensor. Usually is the firmware bundled in the camera that drives the cooling system (since the exact parameter are best/only known to the manufacturer) according to the available current. I'm not aware of APT being able of changing the rate of the cooling, at least for the cameras I use. I can't think of any commercially available camera with such a fast cooling down ramp to jeopardise the structural integrity of the sensor.

Hi Stuart I certainly wouldn't worry about cooling too quickly - as Bogdan says it's unlikely the camera will be able to cool so fast that it cracks the sensor. However, I always use the warming aid in APT (or NINA). Apparently it is possible, if you've cooled to the full 40 degrees below ambient, that when power is cut to the cooler the sensor could experience thermal shock as the sensor temp rises quickly to ambient. Warming happens much faster than cooling if there's a big temp difference. I set the warming target to zero as I don't tend to get too many nights that are above 8 or 9 degrees anyway.

Padraic Moran:
Hi Stuart I certainly wouldn't worry about cooling too quickly - as Bogdan says it's unlikely the camera will be able to cool so fast that it cracks the sensor. However, I always use the warming aid in APT (or NINA). Apparently it is possible, if you've cooled to the full 40 degrees below ambient, that when power is cut to the cooler the sensor could experience thermal shock as the sensor temp rises quickly to ambient. Warming happens much faster than cooling if there's a big temp difference. I set the warming target to zero as I don't tend to get too many nights that are above 8 or 9 degrees anyway.

For a possible fracture, it's not important how fast you cool the sensor but how strong the temperature gradient within the sensor is. If the Peltier element is powerful, it could quickly achieve a very low surface temperature such that the sensor faces -20C on one side while the other is still at room temperature. These are the cases where tensions occur and possible damage. That's to give some background about the theory and why manufacturers mention it.

Usually drivers will take care of it. If I let my Atik do automatic cooling, it takes about 5 minutes to achieve target temperature. However, since there is no hurry at any moment of time, I tell the software the target temperature and set it to cool within 15 or sometimes 20 minutes.

Cheers,
Björn

Think the main issue with cooling too fast is not so much damaging the sensor but getting ice crystals on the sensor window. Which is very rare but can happen. Also important that you do not necessarily need to cool to -20 or something in that range. Most sensor dont benefit very much from going below a certain temp before any dark current gain impact is minimized. Also saves on the darks library if you pick a temperature you can reach throughout the whole year, both summer and winter.

Björn Arnold:
For a possible fracture, it's not important how fast you cool the sensor but how strong the temperature gradient within the sensor is. If the Peltier element is powerful, it could quickly achieve a very low surface temperature such that the sensor faces -20C on one side while the other is still at room temperature. These are the cases where tensions occur and possible damage. That's to give some background about the theory and why manufacturers mention it.

Frankly, neither that is important (despite being nearly impossible to achieve in the real world of consumer cameras). What is important is whether the expansion or contraction of the ceramic substrate is free or not. If it is free no stress applies.

Why is nearly impossible to achieve? Because to cool down the sensor you would need to cool down the cold finger first, while at the same time pumping out the heat by blowing the fan on the heat dissipation fins. Which takes time as all the components have finite heat capacity, finite conductivity and extended lengths. By the time you drop 1 degree below room temperature on the face of the sensor in contact with the cold finger the top of the sensor is already dropping in temperature (in relation to the magnitude of the thermal conductivity of the ceramic substrate and in time, to the total heat capacity, assuming worse case condition of the outer surface being adiabatic). As it progresses, the further drop in temperature is spread across the rather thin substrate (clue: rather more than less) rather than being confined to the inner surface. The only way to achieve that kind of effect you suggest is to apply a thermal cold shock (maybe liquid CO2) to that face, and that would only last an instant.

Hi Stuart,

The camera will only cool down as fast as the camera will allow. As for warming, it’s not good to just turn off the cooler when your done for the night. Depending on weather conditions, temp, humidly, could cause dew to form on the sensor if you just turn it off, that’s why the software has the option that allows you to warm up the chip before shutdown. I usually wait tell cooler power and chip reads 0. As for setting a specific temp to hit and letting it just go, that’s not always a good idea if it‘s not cold enough outside to hit your set-point.

This has been the standard that I was told is best to follow, set your target temp and watch your cooler power, most software will let you see how much power is being used to cool down the chip. Cooler power will usually start at 100% and as it starts to get close to the set-point, cooler power will start to drop. If my cooler power doesn’t stabilize under 80% then I’ll drop it down. In summer I can hit -25C if it’s under 70 degrees and run under 80%. You don’t want to image through the night with the camera running at 100% cooler power. The less power the camera is drawing to stay cool is better for everything running.

Has anyone here really experienced the problems mentioned in practice? Or are these all just theoretical considerations?

Stuart Taylor:
APT has a cooling (and warming) aid which allows you to change the temperature in specific stages with delays in between. I am told that too rapid a change can damage the sensor.

Hi Stuart,

I’m curious, just who and where did you hear that too fast ‘rapidly’ temp change is bad? This is a first for me. The chip can only cool as fast as it can, you can’t force it to cool faster.

And I stated above, the only thing that is not good to do is to just turn off the cooler when it’s been at temp, as it may cause moisture ‘dew’ to form on the sensor, let it warm up slowly, but it can't damage the sensor.

Bob Lockwood:
And I stared above, the only thing that is not good to do is to just turn off the cooler when it’s been at temp, as it may cause moisture ‘dew’ to form on the sensor, let it warm up slowly, but it can't damage the sensor.

I've never ever seen dew forming on the sensor when shutting down, always when powering up. Besides, once shut down who cares. Incidentally, the process is essentially slow as the fan (if it has one) won't be running any more and therefore everything is left to conduction and natural convection (on the outside). I can't think of anything gentler...

Cooling 5minutes to -15C. Starting usually at +10, +20C.
Warming - just power off the camera.
QHY695A.

Well, i was thinking about it too much as well, before or in the past i had only 1 cooled camera, now i have 4, by the time and experience i found out that it is not a big deal, first i always make sure to calculate the ambient and then trying to make the target cooling degree low enough without reaching or exceeding the allowance cooling of the camera, so if it says 40 degree from ambient and it is 30C outside i don't cool down to -10C, i just cool it down either to zero or -5C.

I try to keep cooling time between 3-5 minutes, it is easy to burn out this time somewhere else, but my main problem is when i cool it down to target then for no reason the camera disconnect or i use another app for the camera and back to imaging and the cooling is off, then i don't know if i must cool it down again in 3-5 minutes or i just shrink it to 1 min, i tested and i have no issues, but how can i know if no issues until something bad will happen later and i didn't prevent it from happening from beginning, if i don't feel pain it doesn't mean that i have no injury until i am bleeding a lot and too late, so i hope no shock or accident happens to my cameras, and warming i still don't know, sometimes i let it warm up naturally, and sometimes i just turn it off immediately.

Bob Lockwood:
Stuart Taylor:
APT has a cooling (and warming) aid which allows you to change the temperature in specific stages with delays in between. I am told that too rapid a change can damage the sensor.

Hi Stuart,

I’m curious, just who and where did you hear that too fast ‘rapidly’ temp change is bad? This is a first for me. The chip can only cool as fast as it can, you can’t force it to cool faster.

And I stated above, the only thing that is not good to do is to just turn off the cooler when it’s been at temp, as it may cause moisture ‘dew’ to form on the sensor, let it warm up slowly, but it can't damage the sensor.

Bob, it's a message in APT (which I am using). It warns of thermal shock.
But from what you are all saying, it sounds like even letting it warm up naturally from say -15 up to ambient isn't going to be a problem. So I shall stop worrying

As for dewing up, I think the sensor on the 2600MC is in a nitrogen envelope or something behind the IR cut filter and so isn't susceptible to dewing anyway. (but I am sure someone will correct me if I'm wrong)

Gosh, I'm seeing a lot of "armchair" logic in these responses. Not so much, the knowledge supplied by the manufacturer of the cameras.

For my QHY camera, the manufacturer specifically states that rapid cooling (can't remember about warming also) can result in the damage of the "cooling unit". You break that and you are back to your SLR. Yet everyone giving "conventional wisdom" responses all gravitated to the sensor being the sensitive area.

Also, at least with QHY, they specifically state that YOU should control these rates with the software you use. That means at least for this manufacturer there is no internal safety control in the camera. So please always try to go to the source for this information, and not solely on forums.

Finally, my personal experience is that I have never fried my cooler units or sensors, even when I have been forced to disconnect the cooled camera because of a software hang or computer blue screen event. I will have to say that when this has happened, my QHY camera has warmed only a few degrees during the time it has disconnected and I was able to reconnect. (A minute or two.) This is surprising because a Peltier cooling element is usually quite a thin piece of metal and the sensor not much thicker, if, so you would think the surrounding heat would penetrate rapidly. My ZWO camera heated faster in such a situation. But I had also switched controlling software at the same time. It may also be that the QHY camera employs a thicker cooler plate, or adds more metal to increase the thermal inertial of its setup. BUT, that is all speculation on my part!

Unless you are in a rush, get solid info from the manufacturer and treat your camera thusly. With 15 different opinions from 15 different people, which one would you choose anyway?

Stuart Taylor:
As for dewing up, I think the sensor on the 2600MC is in a nitrogen envelope or something behind the IR cut filter and so isn't susceptible to dewing anyway. (but I am sure someone will correct me if I'm wrong)

Speaking from experience with one ZWO camera, the sensor chamber, behind the AR window is not perfectly sealed. There will come a time that you will begin to see fogging and frosting directly on you sensor. Then you will find the ZWO tutorial on how to open the chamber, pull out the ceramic dessicant tablets and recharge them. These cameras do have gaskets to seal the chamber, but they are not perfect. ZWO specifically advised me to seal the gasket with a bit of silicone grease or cement.

BTW, you will also learn that the cmos sensor surface is not exposed at all. It resides just below a glass cover slip that is directly glued and sealed around the sensor edges. I assume this is by the chip manufacturer, but not sure. So the frost or dew never settles on the cmos surface itself. So don't panic! Also the front chamber window in your camera is heated, so it would never fog. Just touching the front of the camera after a few minutes will tell you that! And when you finally break into that chamber to refresh the dessicant, you will be impressed with how big the heating element is!

In most cases, I think the heated window should be unnecessary. The windows in these cameras are pretty far from the sensor and it would be quite some feat for the cooled sensor (where the cooling physically tightly attached between the sensor and Peltier plate, attached behind) to cause the whole sensor chamber and the huge, thick aluminum window housing to cool significantly. There is just no way. The coolers struggle enough to cool a sensor, let alone many grams of aluminum 10 mm away.

Alan Brunelle:
Gosh, I'm seeing a lot of "armchair" logic in these responses. Not so much, the knowledge supplied by the manufacturer of the cameras.

I guess I am one them, then. I built 2 CCD cameras (with Peltier cooling) in the early days so what can I say... Well I was sitting most of the time but not in an armchair if I recall right. Let's others; SX doesn't say a thing but these were early days and the cooling was blind, so to say. ZWO doesn't say a thing either. The firmware control the rate of cooling as it should, as function of the set temperature and the current rate they decided it was OK to handle. Apparently they aren't bothered either, in which case I'm in good company.

andrea tasselli:
Alan Brunelle:
Gosh, I'm seeing a lot of "armchair" logic in these responses. Not so much, the knowledge supplied by the manufacturer of the cameras.

I guess I am one them, then. I built 2 CCD cameras (with Peltier cooling) in the early days so what can I say... Well I was sitting most of the time but not in an armchair if I recall right. Let's others; SX doesn't say a thing but these were early days and the cooling was blind, so to say. ZWO doesn't say a thing either. The firmware control the rate of cooling as it should, as function of the set temperature and the current rate they decided it was OK to handle. Apparently they aren't bothered either, in which case I'm in good company.

*Andrea, I and certainly no one else can know what the expertise of thecresonders to this or any forum is unless stated outright in the forum. And respectfully, your experience in building ccd cameras with Peltier cooling is not a certain match to any particular brand of camera unless you were on the team of developers. If so, this is important info to state upfront before we send off a first-time owner to break their camera with unsubstantiated advice, experience aside. (Meaning, yes, some have successfully gotten away from heeding the warnings on rapid cooling and warming. That is not proof that such actions cannot damage a camera.)

In my response I gave a direct response to what one manufacturer states about their camera and what their engineers believe are the issues. It may be true that ZWO fails to do so, but ZWO really does respond to requests, and that is the most sensible, simple route to take in Stuart learning about the specifics of his new camera.

Without credentials and specific references, it is always safest to take feedback on forums of this nature with skepticism. Even from those who might be percieved to have much experience. Those who surely know facts will give that info and/or data.

Alan

Of slow cooling and slow warming, unless we are in a race against the time which is not uncommon.
I've lost the guide star several time one night last year while watching a movie but I was just thinking it were passing clouds. Just when I got upstairs I realized that it was nearly starting raining so I disconnected everything in a click and put inside the house.

*Roberto,

I think this happens to all of us and is not uncommon. For me it always seems to come from computer problem. But as I gave my experience with my QHY, its "apparent" slow warming rate may mean warm-up is never an issue for this camera. But when the same happens with my ZWO, warm-up is much faster.

Not to say that the occasional uncontrolled warming cannot damage a camera, the biggest problems may result for those who use this as their only means to warm their camera, or choose 0 min, to cool their camera.

Nothing wrong to ask the manufacturer. Right?

Thanks everyone. I have now asked ZWO and they say there is no concern

Is it because you use APT and there is this statement about the sensor damage? Then you don't need to worry about this, we did not announce anything related officially.

At the risk of repeating myself here, has anyone actually had thermal damage to their camera?

Hello Stuart,
1st there is a lot of good info here in this thread about cooling and warming of the camera sensor, especially about the ZWO brand with the desiccant tables. I have 2 and have to replace or recharge/dry them out 1-2 times a year.

From what I know and have learned here is back in the “old days” when CCD cameras were first coming out sensor shock was a common thing and to help with that you ran thru a slower cooling and warming cycle. With the modern cameras now a days the sensor shock isn’t as common because the software controls the cooling of the camera. This is kinda what many here are saying.

As far as warming goes, before you shut down for the night change your temp by 10º and let it warm as you pack up and put things away. If you want then change it again another 5 or 10º and then disconnect after that.

Just use common sense and you will be fine.

Dale

Olaf Fritsche:
At the risk of repeating myself here, has anyone actually had thermal damage to their camera?

I think might have occurred in cryogenic cameras in the past (professional cameras for deep IR observations), if I recall right but not in the recent past. Amateur-wise, I never heard of such a thing. As I tried explaining in my earlier post such thing is well-nigh as impossible with current amateur technology (Peltier-cooling).

Exactly. And if they do not answer I will go on cooling and warming-up in slow mode

But they did answer. See my post earlier today. They say there is not an issue

Dale Penkala:
1st there is a lot of good info here in this thread about cooling and warming of the camera sensor, especially about the ZWO brand with the desiccant tables. I have 2 and have to replace or recharge/dry them out 1-2 times a year.

yikes! There is no way I am going to be unscrewing my expensive piece of kit and fiddling about with its insides!

Related discussions