ECN Forum
Posted By: Trumpy The Peltier Effect - 11/25/04 12:59 AM
I've looked around the Net until I'm blue in the face.
Could someone please explain to me in English what the Peltier Effect is and how it works?.
Your comments would be most appreciated.

Cheers,
Mike :]

Here's a rather strange website that I found during my searching:
http://www.wordengineering.net/sillycon/
Posted By: hypress Re: The Peltier Effect - 11/25/04 05:46 AM
The Peltier Effect is the voltage generated by dissimilar metals (thermocouple) The effect involves theliberation or absorpotionof heat dependingon the direction of current flow across the junction.
If you know a process instrumentation engineer see if you can borrow their copy of LIPTA'K INSTRUMENT ENGINEERS HANDBOOK (PROCESS MEASUREMENT) there is a good explaination of Peltier Effect.
Posted By: C-H Re: The Peltier Effect - 11/25/04 03:20 PM
In practice, a peltier element is a flat pice of metal with two wires sticking out. If you apply a voltage across the wires, the plate gets hot on one side and cold on the other. The cool thing (no pun intended) is that there are no moving parts in it!

If you put the peltier element between a hot and a cold surface, it will instead output a voltage across the wires.
http://www.heatsink-guide.com/peltier.htm http://www.eureca.de/english/peltierelement/info.html

An excerpt taken from:
http://www.fys.kuleuven.ac.be/pradem/fysici/Peltier.html

Quote

Jean Charles Athanase Peltier (1785 - 1845)
French physicist born in the Somme department of France, Peltier was a watchmaker who gave up his profession at the age of 30 to devote himself to experimental physics. In 1821 T. J. Seebeck had shown that if heat is applied to the junction of a loop of two different conductors a current will be generated. In 1834 Peltier demonstrated the converse effect (the Peltier effect). He found that when a current is passed through a circuit of two different conductors a thermal effect will be found at the junctions. There is a rise or fall in temperature at the junction depending on the direction of current flow.
Posted By: winnie Re: The Peltier Effect - 11/27/04 02:46 PM
The concept that is at the core of any heat pump is 'selection for higher energy particles' at the atomic level.

Heat is simply motion; random motion of particles bouncing against each other, vibrating in place when held by atomic forces, etc. Just atoms, molecules, electrons, etc. bouncing around. Because this _random_ motion, some of the particles will have more kinetic energy than others.

If you could somehow _select_ for the highest energy particles, then the average energy of the particles that remain will be reduced, and the temperature of the particles that remain will go down.

Luckily this segegation of energy levels happens all over the place. For example, when a liquid boils, it is the molecules with the highest energy that escape first. If you were to take a flask of water and pull a sufficient vacuum on it, then the water would boil at room temperature, and get colder.

The transition from liquid to gas represents an 'energy barrier', and only the highest energy particles have sufficient energy to get over this barrier. When particles climb this barrier the heat contained in the remaining material goes down. Similarly, when particles fall down this barrier and condense into the liquid, the heat contained in the fluid increases.

Now, if you simply have a closed container, after some of the liquid evaporates, the volume of the container will become saturated with the vapor, and you will see condensation of vapor back into fluid. The net result is equilibrium, with the same amount of energy being deposited through condensation as is absorbed through evaporation. But by adding a pump to remove the vapor, you can make the cooling process continue.

In an ordinary freon cycle heat pump, you have a region in which freon evaporates, absorbing heat. You then use a pump to remove the vapor and concentrate it elsewhere. The freon condenses on the high pressure side of the pump, adding heat to the fluid on the high pressure side. By selecting pump pressures correctly, you can absorb heat at the desired cold side temperature, and then reject this heat at the desired hot side temperature.

So we see the four pieces of the system: particles climb an energy barrier, selecting for the particles with the highest kinetic energy. The kinetic energy of these particles is converted to potential energy. Then some external energy input is used to move these particles. Then the particles fall down the energy barrier _in a different location_, converting their potential energy back into kinetic energy, and releasing heat. The particles are allowed to cool down to ambient temperature and returned to the evaporating side of things.

A peltier effect device follows this exact same general principal. The particles are no longer freon molecules, but instead individual electrons. In conductors, the random vibrations of electrons are one of the things that carry the heat of the object. By selecting off the highest energy electrons you can lower the temperature of the object.

At any junction of dis-similar conductors, there is a bit of potential energy barrier; electrons going in one direction will have to have slightly higher potential energy, and the electrons that climb this potential energy barrier are the highest energy electrons.

Of course, if you simply have a junction sitting there, as many electrons will fall down the potential wall as climb up it: equilibrium. But apply a voltage, and electrons that happen to make it up the barrier will get tugged away from the junction. Net result: heat absorbed at the junction. Since electrons must flow in a closed circuit, the implication is that there must be at least one other junction someplace else, and that heat will be rejected at that junction.

It gets more complicated when you try to explain why there is a potential energy barrier between conductors, and what happens with 'P-type' semiconductors. I'm still pretty fuzzy on these latter bits. But the essential part is the selection of electrons for highest kinetic energy, and the conversion of that kinetic energy to potential energy as the electron moves through the junction.

-Jon
Posted By: C-H Re: The Peltier Effect - 11/27/04 05:54 PM
Jon,

I like your description. I never understood the Peltier effect at this level.
Posted By: jraef Re: The Peltier Effect - 02/11/05 11:46 PM
Nice explaination.
For further searching, use the search term "Thermo-Electric Cooling". That is the (unofficial?) trade name for systems using peltier effect devices. ITS is their international trade group, and their website hosts a lot of links to other very informative sites. http://www.its.org/
Posted By: Trumpy Re: The Peltier Effect - 02/16/05 02:06 PM
Reason I ask, is because the kiddies, with thier already over-clocked computers are looking for some sort of cooling. [Linked Image]
Posted By: resqcapt19 Re: The Peltier Effect - 02/16/05 04:12 PM
Mike,
Take a look at some of these CPU coolers.
http://www.tomshardware.com/cpu/20010528/
http://forums.extremeoverclocking.com/archive/index.php/f-184.html
http://www.xtremesystems.org/forums/archive/index.php/f-156.html
Posted By: Trumpy Re: The Peltier Effect - 02/22/05 10:25 PM
Hey Don,
Thanks a heap for the links!. [Linked Image]
I see some wierd looking pictures on the site of the first link above.
Especially with bringing water hoses through the sharp edges of a USB port.
© ECN Electrical Forums