I'm not sure if this is the right section to ask this question (or if I should rather have posted it on Physics stack exchange). Anyway, I'm trying to build a cloud chamber that relies on Peltier cell to cool the bottom down. I read online that the needed temperature to make the isopropyl alcohol supersaturated is by reaching a temperature difference where the colder one is about - 25/30 °C. I've tried to use a peltier TEC2 25408 or a TEC1 12708. The first one is suggested by this project (https://www.instructables.com/Peltier-Cooled-Cloud-Chamber/?amp_page=True). To cool the hot side of the Peltier down, I've tried to use a Cooler Master H412R fan, with mediocre results (the temperature I was able to reach was of about - 7°C with the TEC2 and - 11°C with the TEC1). Now I assume the problem is that the fan isn't strong enough and gets overwhelmed by The heat produced by the TEC. I thus wanted to try another kind of fan. As the previous project suggest, I was looking for a Cooler Master 412S, which, as the project I've linked demonstrates, with a TEC2, was able to reach - 40°C. Now I'm a bit skeptical about this results: do you think it is reachable such temperature? And could you recommend me some other efficient fan model?

P.S.: The fan I've talked about, the 412S, is not available on Amazon, do you think that a used one from ebay will do the work? I'm asking this because I don't know if the usage of the fan could slow it down or make it less efficient.

Thank you in advance for your answers.

  • 1
    \$\begingroup\$ Do you have a heat sink for the hot side? Recommending what to buy is off-topic however. \$\endgroup\$
    – Justme
    Commented Mar 7, 2023 at 7:00
  • 1
    \$\begingroup\$ How cold of a temperature you can achieve simply depends on how much heat you can dissipate from the hot side. One of those extremely beefy CPU heatsinks with some powerful fans should work well. The quality of your thermal interface will also matter (what kind of thermal paste/adhesive and how good the bond is, you want the smallest possible temperature gradient across the thermal interface). I imagine that -40C is possible with a large heatsink, powerful fans, and high quality thermal paste. \$\endgroup\$ Commented Mar 7, 2023 at 8:07
  • \$\begingroup\$ A cheap AIO CPU watercooler might be a good way to go for the hot side, since from a mechanical perspective I'd imagine it'll be far more convenient to have the bulky cooler/fan part away from the Peltier element, rather than mounting a huge heatsink to it directly. \$\endgroup\$
    – Polynomial
    Commented Mar 7, 2023 at 8:12
  • \$\begingroup\$ The thermal paste I'm using is the following one (SSY thermal paste) : amazon.it/gp/… \$\endgroup\$
    – Luke__
    Commented Mar 7, 2023 at 11:53

1 Answer 1


It's critical to get the hot side of a Peltier down to the lowest practical temperature to have a good chance of reaching a low temperature on the cold side. Their efficiency plummets when you try to operate with too high a delta T across them.

As it's for an experiment, this allows you to use bulkier equipment than might be sensible for a portable application.

I would recommend using a water block on the hot side. This has a very low thermal resistance, and as another benefit takes much less space around the Peltier than direct air-cooling. You now have several options for cooling the water ...

  • A fan-blown radiator. The radiator can be large, as much as you can budget, which will get the water close to room temperature
  • A bong/swamp/evaporative cooler. This will get your water below room temperature, approaching the dewpoint.
  • Flowing through a bucket of ice, ice being replenished as required from your freezer. This will get the water down to near zero.
  • Just have a large volume of cold water. Water has a tremendous heat capacity and may take a long time to heat a few degrees. This is the easiest way to do a short run experiment, before you have one of the other better methods set up.

It's important to run your Peltier on smooth DC, from (for instance) an ordinary variable output SMPS. You need to minimise the waste heat generated (proportional to RMS current) per heat shifted (proportional to mean current). Don't try to PWM it, as that increases the RMS/mean ratio, making it much less efficient. Don't switch it on/off to control the temperature. This is not only as inefficient as PWM, but also thermally cycles the ceramics, accelerating bond failure.

  • \$\begingroup\$ Thank you very much for your answer. One more question: it doesn't matter the temperature in itself at which the water inside the cooler is, does it? What's important should be the rate at which the water is cooled that is the dissipato on power of the system, if I am not mistaken... \$\endgroup\$
    – Luke__
    Commented Mar 7, 2023 at 12:42
  • \$\begingroup\$ @Luke__ Try reading my very first paragraph again. It matters critically what temperature the cooling water is. Every degree you can reduce the temperarture of the hot side is a 'free' degree colder on the cold side. As you increase the flow rate, you get the peltier closer to the temperature of the incoming water. However, for typical water block and peltier sizes, you quickly reach a limit where it's not much point going above a 'reasonable' flow rate. When you've reached that adequate flow rate, reducing the temperature of the incoming water helps. High delta T across a peltier is bad. \$\endgroup\$
    – Neil_UK
    Commented Mar 7, 2023 at 13:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.