Slashdot Mirror
Thermoelectric Generator With No Moving Parts
Savage-Rabbit writes "These guys have produced a working prototype of a thermoelectric generator. The thing uses extremely cold and hot liquids to achieve a heat transfer through a semiconducting material. This produces a voltage in the semiconductor who can produce up to 50-100 Watts which is actually enough for this thing to have practical uses. This generator could for example be useful in the chemical industry where many production processes generate a lot of excess heat that normally is simply lost. With a thermoelectric generator some of that lost energy could be recovered."
It's an Icelandic company, so they're developing in an environment where there's plentiful hot water (geothermal). I'm wondering if you could get enough hot water from rooftop water-heating panels. Not the thoroughly expensive photo-voltaic, the much simpler black pipes full of water sort. They're surprisingly efficient even in colder climes: my in-laws in the North of England get most of their hot water from a set. The combination might make this very useful for isolated buildings, even outside of volcanic hot zones.
This is just a Peltier device in reverse. Instead of using a forced flow of electrons to drive heat from a cold surface to a warmer one, it is using the flow of heat from a warm surface, through the Peltier element, to a colder surface to drive electrons, generating current.
Seriously, at my university, thermocouples are covered in a sophomore year mechanical engineering class and lab.
This generator could for example be useful in the chemical industry where many production processes generate a lot of excess heat that normally is simply lost.
Not to mention all the heat lost in even more common things such automobile engines.
Actually the flow of heat from one body to another as the two move twords thermal equilibrium can be used to do work. In the device, heat is moving from an area of high tempurature, to one with a lower temperature, not the other way around.
I know lots of things. Most of them are wrong.
I figure, if they used cool water from the Hudson for the cold side, and warm, um, liquid from the urinals for the hot side, they could get the lighting for night games for free!
Though they might want to lower the prices on soda and beer, just to keep things flowing.
Hopefully it could also be used in satalites. I always hear about the extreme temperature differences between the side that faces the sun and the other side.
Welcome to the land of the free...pay toll ahead...no photography...please open your bag...
Does this mean I can water cool my Athlon and keep the computer powered off the waste heat?
Bleh!
Read that closer. They claim 70-80% of the theoretical Carnot maximum efficiency. The max efficiency of a carnot cycle is found from (Th-Tc)/Th, where Th is the temp. of the heat sink, and Tc is the temp of the cold sink (all temperatures in absolute temperature, like Kelvin). That max carnot efficiency is a thermodynamic law, nothing can beat it.
If they can get 70-80% of carnot, that would be great, but it isn't clear how these devices will scale - meaning they might get a great efficiency but low total power. It also isn't clear how hot they will operate (assuming the cold sink is ambient). If the temperature gradient is only a few K, goodbye efficiency (see equation above).
I made a passing glance at their website, and I didn't see an operating temperature, but I didn't look too hard.
-Looking for a job as a materials chemist or multivariat
They claim:
a generator that gets 3 litres per minute (0.8GPM) of 75C (167F) hot water, gives about 50 Watts when also supplied with the same flow of cold water used for cooling.
They also claim to produce about a 20C drop in temperature. Theoretically that's 4,200 Watts (it takes a lot of energy to raise a liter of water 1 degree). So their efficiency is only 1%.
I hope I've done the math right; high school chemistry was half a lifetime ago.
What is the efficiency of the process? A Peltier is around 5% efficient (very wasteful). These guys [powerchips.gi] are working on a similar device and are proclaiming efficiencies in the 70 to 80% range. Consider that a car engine is 15% efficient (approx) or a gas turbine is 30% efficient (approx).
Bear in mind that efficiency isn't necessarily symmetrical. If I'm trying to generate electricity from a 10:1 heat differential across a boundary, I can be up to 90% efficient ((Th-Tc)/Th). But if I'm trying to enforce a 10:1 heat gradient (i.e. keep the cold side cool), I can be at most 11% efficient (Tc/(Th-Tc)).
I'll still only believe PowerChips' numbers when I see a working device with that efficiency, of course.
This generator could for example be useful in the chemical industry where many production processes generate a lot of excess heat that normally is simply lost.
I have a request. I need something that works with body temp and here's what I'd do:
Flip the power breakers off the night of my honey's favorite movie and tell her that the backup generator works off body heat. Oh course it'd be my luck that she'd tell me to start doing jumping jacks....
"Only one thing, is impossible for god: to find any sense in any copyright law on the planet." Mark Twain
These guys have nothing new. If they really wanted something to crow about, they'd produce something like a small vapor turbine running on butane and try to get 12% efficiency out of the thing. If they could spin one of those on fluid bearings a la the people making microturbine generators, it should be just as reliable and quiet.
Scientists restrict study to entire physical universe; creationist
Solar PV panels run up to about 14% efficiency, whereas Peltier junction devices run maybe 5%? If you're able to build 3 times as much area of collector for the same price, and you're willing to put up with the efficiency going up and down along with the rate of heat input (output voltage is proportional to the temperature difference, so you will have very low voltage and thus low power when the sun is low in the sky)... I suppose it might be worth it. I suspect (educated guesswork) that if you run the numbers you'll find that other approaches are more worthwhile.
Scientists restrict study to entire physical universe; creationist
is the term you are looking for. Fuel is apparently still too cheap to make them worthwhile for things like cars; people would rather pay for the extra gas. AFAIK, even heavy trucks are still not using turbocompounding to squeeze the extra few percent out of their diesels. This is odd, because I read about Caterpillar designing a near-adiabatic diesel with turbocompounding around a decade ago, yett there's nothing on the market (but at least they're talking about it).
Scientists restrict study to entire physical universe; creationist
But the temperature of the cold reservoir there is much lower than that attainable on Earth, since the heat is being radiated off into space. The ideal efficiency is therefore, I would guess, something better than 50% (I'd have to know the temperature produced by the radiator system to get the exact ideal efficiency). So the Voyager RTGs would operate somewhere between ~20% and ~50% of the ideal efficiency, which really isn't that bad.
Bugrit! Millenium hand and shrimp!
The Second Law prohibits the transfer of useful work from a high entropy source (heat) into a low entropy source (an electric current).
So the steam turbines at the (nuclear or coal fired) power station aren't powering that computer you're posting from? I guess we all must live near hydroelectric stations then.
These have been around for years. This is just the reverse of a Peltier cooler. Havent you all ever heard of thermocouples?
If you don't understand any of my sayings, come to me in private and I shall take you in my German mouth.
how did this get moded offtopic, it's funny...
"I'll have a Guinness, no wait, make that a Coors Light" -Grad student I work with, who shall remain anonymous...