Peltier Cooling

[Visionary (David)]

Today I finally got around to building a Peltier cooler for my Astro camera. The camera is an Orion deep-sky-colour, 11 model. I was shocked to discover just how much the cooler drew. At 12v it running at about 3-4amps much heavier power consumption than I initially thought the thing would draw.
My battery is rated at about 4amp hours, adding all the other elements I run it means that I will be needing an addition battery.
Understanding that many Astronomer's plays around with Peltier devices, how effective is the Peltier for Astrophotography? Is there any way to configure the device so it draws less current?
In addition, on the Uber-Highend Astro gear the ability to drop temp against ambient temp is cited, are these Uber devices using Peltiers or is there some sophisticated "Uber" Science? Any help in this matter will be accepted with gratitude.

[Shiraz (Ray)]

I found them to be odd devices to use - a few thoughts that might possibly help if you are using a lot of power (although 3-4 amps at 12v is probably reasonable for a 50+ watt device):
1. drop the voltage - they will work from quite low voltages and although they cannot pump as much heat at low voltages, at least they do not generate much heat internally - you will probably find an optimum beyond which more power does not result in more cooling.
2. the heatsinks must be very efficient if you want to transfer lots of heat at high current - if you cannot move the heat away effectively from the output side, you will not get much efficiency and could end up with nothing but a heating element with one side less hot than the other.
3. keep the two sides as thermally isolated as you can - some sealants/cements/heatsink compound etc will conduct heat well enough to partially "short out" the heat pumping activity and any metal screws etc can do the same.
4. use only as much heatsink compound as required to just establish full contact between mating surfaces - I overdid it and found that there was reduced heat transfer and compromised thermal efficiency as excess compound oozed out and transferred heat between the two surface.

as far as I know, the camera manufacturers do not use anything all that special - maybe stacked Peltiers to pump down to fairly low temps, but that is about it. Of course, they are only trying to drop the temperature of a smallish chip and they probably have it in an Argon filled chamber, so the heat they must pump is not excessive most of the time.

[Visionary (David)]

Originally Posted by Shiraz

I found them to be odd devices to use - a few thoughts that might possibly help if you are using a lot of power (although 3-4 amps at 12v is probably reasonable for a 50+ watt device):
1. drop the voltage - they will work from quite low voltages and although they cannot pump as much heat at low voltages, at least they do not generate much heat internally - you will probably find an optimum beyond which more power does not result in more cooling.
2. the heatsinks must be very efficient if you want to transfer lots of heat at high current - if you cannot move the heat away effectively from the output side, you will not get much efficiency and could end up with nothing but a heating element with one side less hot than the other.
3. keep the two sides as thermally isolated as you can - some sealants/cements/heatsink compound etc will conduct heat well enough to partially "short out" the heat pumping activity and any metal screws etc can do the same.
4. use only as much heatsink compound as required to just establish full contact between mating surfaces - I overdid it and found that there was reduced heat transfer and compromised thermal efficiency as excess compound oozed out and transferred heat between the two surface.

as far as I know, the camera manufacturers do not use anything all that special - maybe stacked Peltiers to pump down to fairly low temps, but that is about it. Of course, they are only trying to drop the temperature of a smallish chip and they probably have it in an Argon filled chamber, so the heat they must pump is not excessive most of the time.

That's very helpful! I have a becnch power supply do varying the voltage isn't to difficult. On reflection I am glad I didn't sink a lot of coin into this project, but then again it could well be the cheap Ebay Peltier that is at fault.
When I had the thing running it quite literally sucked the heat out of the camera. Prior to installation the thing got so cold it that on contact it felt like my fingers were burning! They do get very cold very quickly.
The other interesting effect is how the hot air is simply "dumped" by the heat sink, fan combo.
I have read some interesting material on passive Peltier's rather than driving them use them as passive devices that generate a minute current, maybe stack the Peltiers, each successivly delivering some current, remembering of course that the minute current generate is heat being used!
Clearly I need to work on thermal isolation, what I have is OK but from what your describing isolation is critical condition.
There is something clearly worth investigating but I have a feeling it's going to take a much better Peltier than my $5.00 FleaBay model.

[glend (Glen)]

There is a large amount of info on making Peltier (or TECs as they are more commonly known) work effectively in camera cooling. On IIS alone the DIY forum has several long running threads on DSLR cooling with cold fingers powered by TECs, worth going through to learn all about driving them and managing setpoint cooling.

[julianh72 (Julian)]

I made a very effective Peltier Cooler for my ZWO astro-camera using a cheap ($2) 12 volt / 60 watt Pelter TEC and a CPU "heat pipe" cooler.

Even though both the TEC and the heat sink fan are rated at 12 volts, I run them off a 6 volt battery supply. (It's actually a 12 volt / 7.2 amp-hour SLA battery running through a 12 v - 6 v "buck converter" - another eBay buy for about $5.)

The TEC will pull 3 to 5 amps at 12 volts. At 6 volts, the heat-sink fan runs more slowly and quietly (and vibration-free), and the TEC is only pulling about 1.5 amps at initial cool-down, and the current drops lower as the TEC reaches thermal equilibrium, so my 7.2 amp-hour battery will last for several hours. I'm getting a Delta-T of about 24 degrees Celsius at the image sensor.


Some info on my build here:
http://julianh72.blogspot.com.au/201...-asi120mc.html
and here:
http://julianh72.blogspot.com.au/201...ng-on-cpu.html

[ZeroID (Brent)]

Like Glen I'm involved (heavily) with the threads in the DIY forum with cooling DSLRs. And I've built a ZWO ASI 120 MC camera cooling system.

The ZWO TEC system drew about 5-6 amps at 12 volts but it got down to almost -20* C. It caused some dew\ice issues which I also resolved but I moved onto the DSLR project instead mainly because of the better control software and cheaper bigger chips in DSLRs. 5-6 Volts only pulls an amp or so but the cooling effect is minimal at best.

Having just about sorted out a system for that I may revisit the ZWO project especially since ZWO Optical have now released cooled 2 stage cameras and software.

At high amps you do need a substantial heatsink. If you don't you do run the hot cold loop as detailed above. I'm using some thermopipe radiators off some old servers from work with a fan attached. At least I can keep my hands warm on a cold night in the Ob.

[Visionary (David)]

Thats very helpful, I will go to the threads and have a read. Driving the device at 12v is like feeding a hungry dragon, it wants more and more pixies'. It makes an interesting study. Many thanks!