Peltier cooling of whole 1100D DSLR

[Gina]

Managed to do some work on it this afternoon. Attached cold fins and fan, Peltier, and hot side cooler with heat conducting paste as appropriate. Made up a connecting block for the fans and connected up to my bench power supply. Seems to work alright but will need some time to reach equilibrium. No problem The Peltier is 12v 50W. The outside fan keeps the heatsink just luke warm but it does need the fan. The whole unit is quite heavy and I wouldn't like to add that much weight plus camera on the end of my draw tube. But this is only a test box for temperature test runs - the final product will be a lot lighter. Finally, I added some polystyrene insulation round the box.

I post a photo later.

[Gina]

I've set APT to take a series of 5 min darks to see how the sensor temperature varies in the cooling box. Initial temperature was 16C but rose fairly quickly and now seems to have settled down at 21C. I haven't yet got a thermometer in the box so my only indication is the end of the box around the Peltier TEC and cooler.

[JamesF]

Stick a 1-wire thermometer inside the case, Gina

James

[Gina]

Stick a 1-wire thermometer inside the case, Gina

James

Yes, good idea

[Gina]

Here's a couple of photos of the completed test cooling box - with and without camera. The camera is spaced away from the bottom of the box to allow cold air circulation. For the testing the box has a lid which is also covered with insulation.

[Gina]

Sensor temperature dropped to 18C by the end of the 2 hour run I'd set up. I presume this was probably because the ambient temperature had dropped. I was in my warm room but I didn't have any heating on. The outer heatsink temperature had dropped and was no longer feeling luke warm. I presume a Peltier device produces a temperature difference so having a cooler hot side will give a cooler cold side. Maybe more cooling might be obtained by stacking two Peltier devices one on top of the other.

[malc-c]

Gina, Not exactly the same thing, but you might find this interesting

DSLR dry ice cold box

Uses a box similar to yours, but assists the cooling using dry ice

[malc-c]

Gina, I think the reason you are not getting a decent level of cooling is due to the thickness of the ali box you are using. This example uses a hand crafted box from thin ali sheet

Peltier Cooling of Modified Canon Digital Rebel XSi (450D) - Version III -by Gary Honis

[Gina]

Gina, Not exactly the same thing, but you might find this interesting

DSLR dry ice cold box

Uses a box similar to yours, but assists the cooling using dry ice

Interesting Pound and a half of dry ice, eh!? And I thought my Peltier cooler was heavy

[Gina]

Gina, I think the reason you are not getting a decent level of cooling is due to the thickness of the ali box you are using. This example uses a hand crafted box from thin ali sheet

Peltier Cooling of Modified Canon Digital Rebel XSi (450D) - Version III -by Gary Honis

Yes, I've read that and I expect to do much the same for the real thing. I want to get my camera as small as possible. Removing the back would help quite a bit and give better access to the image sensor chip. I may be able to blow cold air right at it Not quite as good as a cold finger but a much less hazardous procedure

[ncjunk]

Sensor temperature dropped to 18C by the end of the 2 hour run I'd set up. I presume this was probably because the ambient temperature had dropped. I was in my warm room but I didn't have any heating on. The outer heatsink temperature had dropped and was no longer feeling luke warm. I presume a Peltier device produces a temperature difference so having a cooler hot side will give a cooler cold side. Maybe more cooling might be obtained by stacking two Peltier devices one on top of the other.

Yes spot on.

Thats why you get two stage peltiers on a lot of astro ccds like these ones-

http://www.ferrotec.com/products/thermal/modules/multiStage/

Enviado desde mi GT-I9003 usando Tapatalk

[Gina]

Ah yes, good I think I might stack 2 Peltiers. A higher hot temperature could be cooled with a smaller heatsink, saving weight.

Been running a test today. Sensor temp started at 16C having taken the camera outdoors to the warm room from the warmer house. It then rose to 23C in about 15m then after a short plateau, started going down again and now after 2 hours is 17C.

The cold box has some condensation on the outside but not frost so it's above freezing. The inside is still quite dry - no condensation so there is no danger of condensation in the camera as it'll be warmer than the box.

[Gina]

Sensor temperature up to 23C now probably with rising ambient in the warm room due to the sunshine. Anyway, with the slowly changing temperature I shall have a good collection of darks. The noise shown in DPP from the CR2 RAW file for 23C with 30m at ISO 3200 shows about 10 noise pixels over the whole frame.

With half hour darks only showing a bit of noise, this camera should be great for DSOs, particularly with cooling.

I shall be working on a new design for the cooling box that I can use on the scope. I have a smaller cooler on order - a round one like Gary uses on the inside of his box. Alternatively, if it gives enough cooling, I might use it on the outside. I want to blow air onto the back of the camera rather than blowing out at the sides. It may be possible to turn the fan over though, depending on the construction.

[Gina]

Camera battery lasted not much more than 3 hours - must sort out a PSU to replace it.

Temperature inside box was 6C not exactly cold! So looks like I shall go for stacking 2 Peltiers. I don't think these cheap Peltiers from eBay are very efficient - probably why they're so cheap Maybe those taken out of cool boxes and drinks coolers are more efficient. At least the difference between sensor and box has been reduced by circulating the cold air. Should be even better with the back off the camera.

With all this messing about with Peltier and fans I get a sensor temperature almost down to ambient - hardly an earth-shattering result Still, it does reduce the sensor temperature by something like 15C with a very significant reduction in noise.

[iwatkins]

This is the problem Gina, TEC devices are not very efficient anyway. They can be very effective but only with big current draws, which I guess you don't want to get into.

Still think a big liquid coolant system might work, would be a faff to set up though. But hey, clear tubes with lime green coolant flowing would look cool

I guess the big question, mains or batteries? If mains then consider a proper phase change cooling, like how a normal fridge or air con system works. Easy to get well below ambient then.

As you'll end up actually be insulating the sensor because of the rest of the camera body around it, I think just cooling the air in the box will struggle to get you where you want.

I've been partially involved with the supercomputers at my work for the past 20 years. Gone from exotic phase change cooling, to water cooling, to cold finger/heatsink (huge), to forced air to purely radiative cooling.

One of the more interesting ones were these super fast small fans, mounted in a Venturi tube focused onto a small heatsink on the chips. Think that was Silicon Graphics/Cray. Kept the chips at ambient from what I remember. The idea was to have very very fast air, but low volume flowing over the heatsinks.

Cheers

Ian

[Gina]

This is the problem Gina, TEC devices are not very efficient anyway. They can be very effective but only with big current draws, which I guess you don't want to get into.

I could cope with quite high currents as I'm planning to use spare PC PSU(s) to provide the power.

Still think a big liquid coolant system might work, would be a faff to set up though. But hey, clear tubes with lime green coolant flowing would look cool

Yes, it would look cool

I guess the big question, mains or batteries? If mains then consider a proper phase change cooling, like how a normal fridge or air con system works. Easy to get well below ambient .then.

It's mains - I have a permanent obsy with mains power. I'll have to look into that - sounds a bit complicated.

As you'll end up actually be insulating the sensor because of the rest of the camera body around it, I think just cooling the air in the box will struggle to get you where you want.

Yes, I agree, it's far from ideal. I may have another look at direct sensor cooling. I might just go for making up one ideal camera from the two bodies I've got. I've never actually dismissed the idea of a cold finger totally. I'm investigating all possibilities. One option is to replace the metal frame that carries the sensor with a fibreglass one and isolating the sensor thermally. Then the hot side of a Peltier device could be connected to the camera frame and other metal parts (eg. filter wheel casing) to both dissipate heat and warm them up to prevent condensation. That would leave just the sensor and it's PCB to be sealed to prevent moisture ingress. I guess quite a small Peltier TEC would be sufficient and this system would save a lot of weight.

I've been partially involved with the supercomputers at my work for the past 20 years. Gone from exotic phase change cooling, to water cooling, to cold finger/heatsink (huge), to forced air to purely radiative cooling.

That's very interesting If I could just get the sensor to run at ambient temperature by conducting the heat away efficiently, that would probably be sufficient.

One of the more interesting ones were these super fast small fans, mounted in a Venturi tube focused onto a small heatsink on the chips. Think that was Silicon Graphics/Cray. Kept the chips at ambient from what I remember. The idea was to have very very fast air, but low volume flowing over the heatsinks.

Ah yes, that's something along the lines of what I've been thinking of for cooling the sensor. A fast flow of cold air over the back of the sensor might be quite effective. Maybe not as good as a cold finger and thermal paste but might be easier to implement. Food for thought. Wonder where you can get small, fast fans...

Thanks for all your thoughts and ideas

[iwatkins]

No worries Gina

As for small powerful fans, the closest you would get would be a ducted fan for RC models. E.g. -ON SALE-90mm Duct Fan + 1750KV Brushless Motor for lipo RC Jet-IN STOCK FREE SH | eBay

Have a look on Youtube for "ducted fan" to see what they can do.

They do have a big amp draw and are not quiet, but they shift air like nothing else their size.

Basically, once you have a method to get heat away from the sensor to at least the interior on the box, there are lot of ways to get it out of the box.

Cheers

Ian

[fwm891]

Gina, I found putting the camera body in the fridge before hand kicks the cooling process off very quickly....

http://stargazerslounge.com/diy-astronomer/147163-chiller-box-dslr-camera-body.html

[rcheshire]

Hi Gina. I've been reading your posts about cooling an 1100D. I found out about your efforts from an Ice In Space member.

Cold fingers are not easy to implement so I'm very interested in how your latest approach is performing.

If you are interested, here is a link to my 1000D cooling mod, which was clearly less painful than your experiments with the 1100D.

Regards,

Rowland.

[Gina]

I'm not using any cooling at present. The night-time temperature is low enough to give reasonably low noise levels. OTOH I have a modified sensor assembly complete with cold finger I expect to try out later.

[yesyes]

If you are interested, here is a link to my 1000D cooling mod, which was clearly less painful than your experiments with the 1100D.

Regards,

Rowland.

Thanks a lot for that link. Very interesting read. It is very close to what I'm planning to do

http://stargazerslounge.com/diy-astronomer/183121-eos-1000d-extreme-modding.html