Magnetic cooling - the new era of refridgerative cooling

[The Pentium Guy]

The magnetocaloric effect takes place when a metal becomes magnetized, it becomes heated. Similarly, when a metal becomes demagnetized, it becomes cooled. An electromagnet with a high powered DC source (aka a power supply) could be used to power the electromagnet, provided that many coils are wrapped around the magnet itself.

Quoted from http://www.sciencenews.org/articles/20040626/fob6.asp:

When a magnetic-refrigerant material is exposed to a magnetic field, the field forces the spins of electrons in the material to align. As a result, the material heats up. Removing the field permits the electrons to relax into less-ordered states, and the material cools down. By cycling the material through these hot and cold states and venting away the heat, the system can generate an overall cooling effect.

During the heating stage, the magnet itself can be cooled using a typical heatsink, but while the magnet is in the cooling stage, a liquid (or even air, if you wish to create an air chiller) flowing through the magnet could be used to circulate the chilled water that can be used in a CPU cooler.

http://vector.iop.org/abstract/0750309229

First of all this is related to practical application of the MCE and magnetic materials in refrigeration devices and, especially, in magnetic refrigerators, which work on magnetic refrigeration cycles instead of conventional vapour gas cycles. Recently a series of acting magnetic refrigerator prototypes have been developed and created. The particular importance is that the created prototypes work at room temperature and have significant potential to be incorporated into the marketplace. Magnetic refrigerators are characterized by compactness, high effectiveness, low energy consumption and environmental safety.

Two of the magnetic coolers could be used to create an alternating effect, using a solenoid to switch between magnets during their cooling cycles.

Relatively new idea, but some US department (department of energy I think) tested this with a 500W power supply and acheived good results. Proper care must be ensured to make sure the magnet won't destroy the adjacent computer parts, therefore an external system might be better.

Homemade tesla-coils could be used to provide this same effect for cooling.

Just wondering what your thoughts are,
-The Pentium Guy

 

[Fullmetal Chocobo]

Sign me up. I'll toss another PSU into my machine for it's testing... That sounds interesting.

 

[scrawnypaleguy]

Too many big words... pics?

 

[tw33ter]

Originally posted by: The Pentium Guy
Proper care must be ensured to make sure the magnet won't destroy the adjacent computer parts, therefore an external system might be better.

how would one do that?

 

[The Pentium Guy]

Sorry for the lack of response, been busy the past week.

What's been done recently is an alloy of gadolinium (in a powdered form for maximum surface area) was taken and was placed in a wheel. The wheel has this one small gap in it, and outside the slit lies a powerful magnet. The gadolinium powder gets heated due to the magnetocaloric effect. As soon as the gadolinium passes the slit, water comes to the rescue and cools it. The gadolinium cools further becuase it is demagnetized (the magnetocaloric effect works both ways). Depending on the strengh of the magnetic field, and the pump you use for the water, it can drop about 50F below room temperature. Another stream of water rushes in and is cooled by the cold plate, and is carried over to the CPU.

Somene asked for pics:
http://img523.imageshack.us/img523/7034/magneticcooling1fi.jpg
Edit:
http://www.external.ameslab.gov/techservices/graphics/refrigwheels.jpg if you want to see a more professional image.

I'm thinking this is way overkill for a CPU cooler, unless you're a hardcore overclocker seeking to replace phase and acheive sub-zero temperatures. Personally, I don't care for that. Instead of using super-expensive gadolinium, iron is a much cheaper alternative and it exhibits 2-3 times less of an electromagnetic effect than the gadolinium alloy (according to the sources I've read). Call me anal, but I prefer keeping liquid out of my system. I was thinking the same approach could work, except with air cooling and two magnets instead of 1.

I'm still amazed at the ingenuity of these scientists that came up with this wheel-with-gap idea, sometimes the simplest solution is the best. Their method uses a permanent magnet that heats a target (the gadolinium powder).

My idea was using an electromagnet and using that as the target directly. Perhaps I should explain in more detail.
I planned on having 2 electromagnets. While one electromagnet was getting magnetized, it would be cooled by a fan (and possibly the electromagnet would have a heatsink on it so you could cool it faster), and once you demagnetize it that cool air could be blown into the heatsink, giving it a boost. The other electromagnet would be heated while one is cooled, giving it a nice alternating effect, so that a constant flow of air goes to the CPU. However, this method's kind of a hitch because I'd have to work with transistors and figure out how to switch the circut on/off and then get the fans blowing at the right time.

I still like that wheel idea, I'd just have to worry about cooling the thing somehow by using air. I'm worried about one thing in particular: electric DC motors are comprised of magnets as well. Won't the strong magnetic field seriously screw up the motor?

Sorry for the long post, but responses would be really appreciated. For my school project, I'm designing a CPU cooler, and I figured that heatsinks were already pretty damn efficient enough, so an air chiller might be a nice addition to a current heatsink.

Thanks,
-The Pentium Guy