How Cold Can I Go? Low Temperatures With a Peltier Stack and a CPU Cooler

Intro: How Cold Can I Go? Low Temperatures With a Peltier Stack and a CPU Cooler

In one of my physics projects, I tried to achieve the lowest possible temperature using simple means. For this I used a CPU cooler and so-called Peltier elements. But how do Peltier elements work?

In physics there is the so-called Seebeck effect, also known as the thermoelectric effect (wikipedia). If you have two different metal wires and bring the two soldering points to different temperatures, a so-called thermal voltage U is measured. This depends on the metal pairing and the temperature difference between the two soldering points. The following therefore applies: U = k * (T1 - T2).

As already mentioned, the factor k depends on the metal used. There is a thermoelectric voltage series in which the metals are ranked according to the voltage they produce. Logically, for the highest possible voltage, two metals are used that are as far apart as possible in the thermoelectric voltage series. But how can this effect be used for cooling? In physics, the principle of reversal often applies. As already mentioned, a temperature difference causes a thermoelectric voltage in the Seebeck effect.

Conversely, if a voltage U is applied to both ends, the two soldering points will have different temperatures. One point of release becomes hot, the other cold. This is the Peltier effect. This effect is used in so-called Peltier elements by providing a large number of soldering points and connecting one plate of the Peltier element to the cold soldering points and the other plate to the hot soldering points. If an electrical voltage U is applied to the Peltier element, a temperature difference between the two plates is obtained.

STEP 1:

In order to achieve the lowest possible temperature, a so-called Peltier stack is used. These are 3-4 Peltier elements arranged one above the other, with the warm underside of the upper Peltier element being cooled by the cold upper side of the lower Peltier element. The performance of the Peltier elements used should decrease towards the top. This is the only way the heat can be sufficiently dissipated. For this purpose, the Peltier elements are also operated with ever lower voltage.

I use a stack consisting of 4 Peltier elements, namely the model 0703 at the top operated with about 0.7V, below the model 3104 operated at 3.3V, then the 7108 with 5V and at the bottom a 12709 at 12V voltage. But here you can experiment according to your wishes and find out which combination can be used to achieve the lowest temperature. The electrical voltages of 3.3V, 5V and 12V are all provided by the converted PC power supply. Another controllable power pack is only required for the 0.7V of the uppermost Peltier element.

The lowest temperature I achieved was -64°C (= -83.2 degrees Fahrenheit). Of course it would have been nice if I had achieved -78.4°C (= -109 degrees fahrenheit). Because then I could even have made dry ice. But maybe one of you can do it.

By the way the lowest possible temperature is the absolute zero temperature at -273°C or 0 Kelvin. You can't go lower. Here is my simple instructable, how to determine this absolute zero temperature: instructable

If you are interested in other exciting physics projects, here is my homepage and my YouTube channel:

Homepage

Youtube-channel

In this sense, have fun experimenting and Eureka...

Comments

[aidanat1]

So if your really trying to maximize cooling you need to look at everything as a system including the TECs as a source of thermal energy transfer. Each TEC should have a lot of insulation around them and a block of aluminum with thermal paste in between them. Layering each TEC with insulation will reduce heat creep and minimize the over all amount of thermal energy they have to move. Obtaining a steady state temperature is the goal hear so keeping it in a well insulated/enclosed environment will help(unless your trying to get an object cold even when exposed to open air). As for with heat dissipation going for a cpu cooler makes sense but, also making sure you get fresh room temp air to cool your aluminum radiator is important. (Maximizing airflow over your tec will help it dissipate and allow your TECs to get even cooler. Related to this, Your TECs max cold temp will be directly dependent on ambient environment temp. Someone doing this test in a 60 degree environment will achieve Colder temps than someone in a 80 degree environment. Another note but, I’m Not sure how cold thermal paste can get but I’m sure condensation on the surface won’t really be beneficial. It likely looses efficiency as it gets colder so another alternative might be better.

[The Mod]

Wait...dry ice is just really cold....huh weird; I can't feel the cold stuff from the mist stuff lol

[mrscience6502]

That's really neat! I tried the same thing a year ago. But I only made it down to -28F. Fascinating!

[stoppi71]

Hi! With a water cooler I assume that you could reach even lower temperatures. Unfortunately my water cooler for this project didn't work, so I had to use my heatpipe-air-cpu-cooler...