Instructables

could you convert heat into electricity?and how?

not much to say.
i need to know if there is a way to turn heat energy into electricity.
almost like a solar panel. (but thats light energy)

but something small, almoust hand held.
i also want to know if you can reverse the effect of a heater?
heat - energy insted of the outher way around.

its for a project.


likewho3 years ago
I’ve tried standard peltier modules for power generation with limited success and very short service life. They cannot hold up to the higher temperatures necessary for good power generation performance. I found a terrific supplier with both standard temperature and high temperature TEG modules designed specifically for power generation. They sell a lot of devices on eBay and you can also buy from them direct. The company is Thermal Enterprises and here is are links to a couple of their eBay items.

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=310148993913&ssPageName=STRK:MESELX:IT


http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=310209561834&ssPageName=STRK:MESELX:IT
interstar4 years ago
This guy uses sound : http://www.unews.utah.edu/p/?r=053007-1
seandogue4 years ago
I'm guessing you want a direct thermoelectric transfer solution.

If not, you could use convective flow to drive a fan blade like those "angels ring the bells" gizmoes" that people put out around Christmas time , (which would in turn drive a generator)
There's more than one way to convert heat into electricity. 

The usual trick is by using a heat engine driving a generator. That is the heat engine to converts heat into mechanical energy, e.g. a spinning shaft, and then the generator converts (most of) the mechanical energy into electricity.  This is the way the pros do it, for example in commercial thermal-electricity generating plants.

Another trick that is sort of coming into vogue, are these solid state Seebeck/Peltier/Thompson devices.  Essentially, one of these devices looks like a lump of metal with two fat wires coming out of it. (See picture.) The way it works is you apply a temperature gradient to it (make one side hot, the other cold), then an EMF will appear across those two wires, and  you can actually get electricity out of it.  It seems like magic, but it makes sense when you consider that these lumps of metal were crafted by elves, and only elves are capable of engineering of such skill and subtlety.

I was looking for someone who sold such elfin magic, for people who live in remote cabins heated by a wood stove, for the purpose of squeezing a few amperes from the wood stove.  Perhaps if Ted Kaczynski had one of these in his cabin, he wouldn't have been such a stick-in-the-mud about technology, and maybe he wouldn't have tried to blow up all those people. Er... Maybe. Anyway, here's the link, but it's basically the first one I found, and I know nothing about this company, other than the observation that these modules are kind of expensive.
http://tegpower.com/products.html

Just for the purposes of your science project, you could probably get away with using one of those Peltier cooling modules, and driving it "backwards".  That is putting a temperature gradient across it, and seeing if this causes any current to flow through the wires.  The reason I suggest this route, is that I expect the Peltier modules (designed for cooling; i.e. heat-pumping) to be a lot cheaper than the modules sold by TegPower, linked above.  Sometimes in thrift stores you can find an old in-your-car tiny cooler/heater/refrigerator, containing a single Peltier module. (See picture.) I think these things find their way into the thrift market, because honestly they don't work that well, not nearly as effective for cooling, or as cheap,  as a styrofoam box with some water-ice in it.

Other methods of converting heat to electricity tend to be more complicated, e.g. using heat to smelt zinc, to use as fuel in a special fuel-cell, or ordinary alkaline battery.

Final note regarding heat as an energy source.  Heat only moves from a place that is hot, to a place that is cooler. Moreover the "speed" or "willingness" with which it moves depends on the magnitude of the temperature difference.  There is a notion of quality, usefulness, or exergy. Usually the higher the temperature difference,the more useful work you can possibly extract.
peltier-cooler-for-yer-car.jpg
Here's a picture with the heatsink removed so that you can see the device itself.  It looks like a white square.
peltier-cooler-wo-heatsink.jpg
jeff-o4 years ago
Well, there are a few ways of doing it.

A small but inefficient way of doing it is with a thermoelectric element called a Seebeck device.  It converts heat directly into electricity.

If you've got a lot of heat to play around with, you could use it to boil a liquid into steam, and use that steam to turn a turbine generator.  You can do it using water, which is what coal, natural gas, and nuclear power plants use.  Or you can use a liquid with a lower boiling point in a sealed system.  The liquid would boil at, say, 50 Celcius and turn a turbine.

Other systems use hundreds of mirrors to focus light on a single point, boiling salt at temperatures of thousands of degrees.  But I dodn't think that's the solution you're after.  ;)
kelseymh4 years ago
Look up "thermoelectricity" and "Peltier effect".  The latter is the inverse process, but physics is reversible :-)
jeff-o kelseymh4 years ago
Actually, the proper term to look up would be the Seebeck effect.  ;)
kelseymh jeff-o4 years ago
Yeah, yeah, picky picky picky.  :-D
lemonie4 years ago
What's your heat source?
Read up on Stirling engines, as another way (plus electrical generator)

L
More efficient for sure, but a heck of a lot more work.
We don't know what they've got in terms of heat, money, tooling. I've seen home made Stirling engines, but it was general in the area of heat-to-power.

L
You use thermocouples, and you use a lot of them. A marginally better way is to use a Peltier cell, but Peltier cells don't work above about 150 C, they melt. Thermocouples will work with one end glowing red.

Typical efficiency though is <2%