About the Conservation of Energy...

Energy cannot be created or destroyed, it can only be changed, as most of us know (Or was that proven wrong? Not sure). Due to that, it can only be changed into another form of energy, where heat is almost, if not always, a by-product, which we generally find useless when heat isn’t what we’re after.

We can pretty much convert almost all energy (Kinetic/Nuclear/Radiant/Chemical/etc.) into electric energy.

So is it possible, even theoretically, to make a device that turns thermal energy into something more useful like electrical energy, which converts more heat than what it would most likely give off? Is there energy that is a by-product of whatever we do that we simply cannot use?

like say, geo-thermal ?

Use a thermoelectric generator, to convert heat into electrical power. They’ve existed for quite some time. It relies on a property called a thermocouple. You’ll be able to pick up a half-dozen or so different sensors using the property, from any good electronics store.

Huh. My physics teacher said that it’s impossible to do that (By that, I mean making the air around the machine colder).

You won’t make the air around the machine colder. I mean you could, you could use a heat exchanger to pull the warm air out, but you’re only dumping it elsewhere. Rather, you’ll use some of the heat energy generated, and convert it into electrical current. You will never get 100% conversion, and waste heat will be created – just not as much as would otherwise be the case.

But could you put that into a loop of some kind? Use another machine to convert some of the heat dumped from the previous machine into energy, so on and so forth?

Sure, but each time you add another machine, that additional machine will produce some additional waste heat that is not converted. It will not be 100% efficient with the output from the one prior, so some of the output from that machine is still heat, plus the output from this next machine.

As you add more and more machines, the surplus heat produced will build and build and build.

Actually you could use the fact that hot objects glow and transfer the electromagnetic radiation that they produce into electricity.

You can use ambient heat to do work (such as creating electricity), but only by transferring it from an area of high heat to an area of low heat, and in so doing you erode that temperature gradient until there is the same temperature everywhere, and no more work can be done.

In a spectacular display of good timing, German researchers have announced the creation of a new type of thermoelectric generator, Camoraz. A printable array version, with many, many of them working in parallel (just like the example you gave) was announced about a month before you created this thread. Found it tonight whilst browsing Franhofer’s research archives.

It won’t lower the temperature around the device, but it will provide internal power for tiny sensors that don’t need any apparent powersource to run electronics. They can be entirely sealed within the device.

you’re confusing and blurring the first and second laws.
you can get an infinite loop, except that ther’es always a leak somewhere in the system. it eventually winds down. eventually. adding another thermopile reduces the thermal gradient available to previous thermopiles.

Originally posted by Camoraz:

But could you put that into a loop of some kind? Use another machine to convert some of the heat dumped from the previous machine into energy, so on and so forth?

You don’t mention the law of diminishing returns, though I’m not sure whether or not this is because you’re focused more on making this hypothetical machine just more efficient and not 100% efficient.

vikaTae and fractalman have the right idea. What you seem seeking is the capture of all the produced energy of a closed system, which is impossible. Energy is not created or destroyed, however it’s more than welcome to dissipate. In that way, per the sensors vikaTae mentioned, it’s inevitable that some of the heat/energy produced will dissipate before it ever reaches a sensor, no matter how close they are. Some of the energy absorbed by the atoms of the medium is also not passed on. As it stands (so far as I know), it’s not possible to prevent this completely.

That being said, the crux of the issue seems to be your desire to bypass entropy. The catch: while it is possible for a closed system to create order, it does so only when there’s a greater increase in disorder elsewhere in the system… :/ I have no way of personally knowing whether or not it’s possible to create a system as you describe with al but perfect energy (re)capture, but, if so, when all the numbers have been crunched you’ll still be in the red.