0
\$\begingroup\$

I'm building a project based on ltc3108, Linear Technologies IC, and the Peltier cells, TEC 12706. The basic idea is to recharge a battery by the temperature difference between the body-skin and the room. The problem is the current, how many cells do I need to have a proper current? Is it best to connect them in series or in parallel?

Thanks :-)

\$\endgroup\$
  • \$\begingroup\$ I suggest that you watch this EEVBlog video youtube.com/watch?v=y4OeOQtiW0w about a peltier-skin heated smart watch. Dave explains why the available power is BY FAR not enough to power the watch. All in all, the amount of power generated that way is simply not worth the trouble. And as Olin answers: the current is irrelevant, 1000 A at 0 V is still zero power. What matters is the amount of power you can harvest. Even a small solar panel the size of the peltier element is bound to produce more power. \$\endgroup\$ – Bimpelrekkie Sep 5 '17 at 12:53
4
\$\begingroup\$

The problem is way more than just "the current".

Before you start this project, look up something called the Carnot efficiency. That's a bit of basic physics that tells you the maximum efficiency you can get power from a temperature difference with.

Let's say the body temperature is 37 °C and the room is 21 °C. That's about 310 °K and 294 °K. Even a ideal heat engine can only extract about 5% of the available power. Maybe you can actually realize 1%. Now figure a human runs at about 50 W (being generous) just sitting around at 21 °C. That means you'd get 500 mW if every last bit of heat from the human went thru your converters before being released into the room.

Unless you are prepared for miniscule amounts of power and/or a cumbersome experience for the human, this scheme isn't going to work.

\$\endgroup\$
  • 1
    \$\begingroup\$ I would add that human skin is about 2 square meters, thats 250mW/sqm or 250nW/sqmm. TEGs are good for a few uW, not more. \$\endgroup\$ – Vladimir Cravero Sep 5 '17 at 13:49
0
\$\begingroup\$

The best TEGs currently produced have a power efficiency of about 2-3%, measured at large temperature gradients. E.g. Tellurex quotes an efficiency of 3% for a ΔT of 75°C. A wearable gadget will have a ΔT of 5°C max, unless you plan to actively blow air on it. So you can expect to harvest 0,2% of the power produced by a human body, which would be about 0,1W, if you cover the body with TEGs completely. Covering 1% of the body (which is above the limit of a gadget what I'd consider wearing, but not unrealistic) will give you 1mW of power, about as much as a solar-powered calculator requires.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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