Hacker News Reader

Comment by dragontamer | Hacker News Reader

I was nerdsniped by a coworker, who was talking about personal projects. She was working on an wine-rack / cabinet that held a specific temperature. Key to the design is a Peltier module to pump the temperature to the ideal for wine storage.

CUI's application note on Peltier modules was very well written and a joy to read. I've barely touched on the physics of this subject, but it was so clearly written that I was able to understand the overall design process to use this kind of device.

Overall, Peltier modules are far less efficient than classic "compressor" refrigeration units. But Peltier modules are far smaller and far simpler to operate: just apply voltage, and away you go. As such, Peltier modules are the ideal easy-to-use tool for hobbyists, even if its not quite as efficient as what the professional HVAC people use.

View on HN

Just to build on your comment so other hobbyists don't get the wrong idea, Peltier modules are significantly less efficient than compressor-based coolers. To the tune of 30% or more efficiency drop depending on ambient conditions and your heatsink setup. It's also important to note that the overall setup for a Peltier system ends up being a lot bulkier than most people expect when they start. You need a really beefy hot side radiator and fan to get any significant cooling potential out of the system - think large CPU heatsink + fan size and you're close. If it's humid you'll also have to deal with condensation on the cold side so you don't accidentally get water dripping on your electronics.

I agree. Peltier fits a narrow niche. If you want a small temperature drop and can dissipate the heat on the hot side and are okay with the inefficiency it's for you. I tried to create a Peltier cooling system for a wearable device. It was a miserable failure. WIthout a way to dissipate the heat (your CPU heatsink + fan comment) it resulted in a net heat gain.

Forgetting about the enormous heat sink is a typical rookie mistake. Those things produce lots of heat when you try to cool something. They are much better suited for heating something to a specific temperature than cooling.

Would an aerogel-like material suffice as an 'enormous heat sink'? Has anyone tried this?

Edit: would a metal aerogel not count? (Definitely not maker ready though, I assume)

http://www.aerogel.org/?p=932

Aerogels are extremely good thermal insulators. For a good heat sink you want a really good thermal conductor.

Very much not. They be better at acting as insulation.

A heat sink needs high thermal conductivity. Hence why most heat sinks you see are metal.

A metal aerogel would have very high surface area (much like adsorption humidity control crystals) but that's not really helpful when you then need to move the fluid touching all that surface area in order to remove the heat.

Typically the fluid will be air which sucks at moving heat (hence why aerogels are great insulators as they hold a lot of air).

I'm not positive but I suspect that an aerogel in a higher specific heat fluid wouldn't do any better as the flow volume through the aerogel would be quite low bounded by the volume of the aerogel.

If you want to cool the hotside of a TEC you need to push a lot of watts per unit area. The best things for that are high specific heat fluids which are often used in compressor systems and then they're termed refrigerants so you're back to traditional air conditioning. The other option is very cold fluids (liquid nitrogen) but at that point you may be better off bringing the LN2 into more direct contact with whatever you're trying to cool.

An aerogel would be among the worst things to use as a heatsink. They are mostly air, held in place. Air has a very low thermal conductivity, low density, and low heat capacity. Air cooling is used because air is abundant and usually readily available, and relies on convection, which an aerogel prevents.

A much better solution are heat pipes, which utilize convection internally to create very high effective thermal conductivity. Assuming the system is not in continual use, the need for a large radiator can be avoided, by using a tank of water or oil (if electrical conductivity is a concern) to dump heat into. With a tank of water, you can also cool below ambient temperature by using evaporative cooling.

One advantage that they have is that they can pump the heat in both directions and change direction very quickly. This means that you can keep temperature very precisely with a closed regulation loop.

Oh yeah, definitely keep an eye on how big practical peltier modules are, so you get a basic gist of how to use these things: https://www.adafruit.com/product/1335

But while most people are talking about Peltier as a cooling module, note that it can be used as a more efficient heater. You can spend 36W of power (12V at 3A), and move maybe 10W. That's a total of 46W of Heat!

In essence: a Peltier is always going to be a more efficient heating device than a standard coil. Heating would be sufficient for tasks like Plant Seed germination or Bread Proofing... seeds / bread want to be around 90F or 32C, so you'd need to create a 110F / 43C source of heat.

Classically, people used light-bulbs to heat dough or seeds for germination, but a Peltier would be easier to control and more efficient.

-----------

Refrigeration is cool (hah, punny), but precise heating also has a lot of good uses... even at the only ~20C or 36F differential where Peltiers are practical. Going +/- 20C or 36F is still a useful trick.

Reply on news.ycombinator.com