Sciencemadness Discussion Board

Peltier Junctions for electricity ?

noxx - 18-1-2008 at 16:37

Hey guys,
I wasn't able to sleep yesterday and I was thinking about something...
Is it possible to generate electricity with a peltier junction with one side painted in black and with a magnifying glass to magnify the sun and the other side with an heat sink

This device could be used outside here in Quebec in winter where we get -10°C to -25°C (outside of course)

Do someone knows the output of peltier junctions when you use them to generate electricity ??

Thanks a lot.



[Edited on 18-1-2008 by noxx]

-jeffB - 18-1-2008 at 16:50

Quote:
Originally posted by noxx
Hey guys,
I wasn't able to sleep yesterday and I was thinking about something...
Is it possible to generate electricity with a peltier junction with one side painted in black and with a magnifying glass to magnify the sun and the other side with an heat sink

This device could be used outside here in Quebec in winter where we get -10°C to -25°C (outside of course)
[Edited on 18-1-2008 by noxx]


Sure, but the output is tiny. You're limited to Carnot efficiency, which with a -10C cold side and 80C warm side (a reasonable safe limit for hot-junction temperature) works out to 1-(263/353) = about 25%. In practice, I doubt you'd see as much as 10% efficiency.

Xenoid - 18-1-2008 at 17:25

When used in this mode they are termed thermoelectric generators.

There are many websites devoted to thermoelectric generators. There is a list at the bottom of this page;

http://www.peltier-info.com/generators.html

In general these devices run on gas burners, so you are looking at temperature differentials of several hundred degrees!

Here is a general page on peltier devices;

http://www.peltier-info.com/

For a historical perspective this page is quite interesting, dealing with thermopiles and the high voltages that can be developed by connecting junctions in series;

http://www.dself.dsl.pipex.com/MUSEUM/POWER/thermoelectric/t...

microcosmicus - 18-1-2008 at 17:58

Consider using a thermocouple which can stand higher temperatures (such as, say,
an iron/constantan thermocouple) --- as long as you are going to use a magnifying glass,
you might as well heat that junction to something like 1000K (red heat) or thereabouts,
to get reasonable efficiency. For instance, think of a Fresnel sheet lens focused on
one of those thermocouples used in stove pilot lights.

In the nineteenth century, batteries thermocouples heated by flames were used for
electroplating and similar applications requiring serious power. More recently,
thermocouples heated by radioactive material were used as power sources in spacecraft.
As long as you operate at a reasonably high temperature, this is a plausible way of
generating electricity.

For reference, here is a page on old thermoelectric generators:

http://www.dself.dsl.pipex.com/MUSEUM/POWER/thermoelectric/t...

Jamjar - 18-1-2008 at 22:48

Wood stove top thermoelectric generator, sink to ambient inside air.

Generating Electricity for Families in Northern Sweden

"Through this project we have shown that it is possible to generate small amounts of electric energy without introducing new resources to the existing environment. The small fan- cooled unit on the wood-fired stove top can generate up to 100 Wh per day without maintenance or noise, enough to feed low-watt fluorescent lights and a small TV."

"the generator will constantly generate 3-9 watts depending on the temperature of the stove top"

497 - 18-1-2008 at 23:23

i have researched thermoelectrics quite a bit, i love the idea of the solid state heat to electricity conversion, but when it comes down to it, they just aren't efficient enough. real world yields would be like 1 to 5 percent if you're lucky (even with a high temperature difference). just the sheer amount of heat you have through it to get any decent amount of power is prohibitive in many cases. also many of the more effective metal combinations like doped bismuth telluride can't handle very high temps. also most halfway decent thermoelectric alloys are pretty damn expensive even if you build the unit yourself. you'd be better off running a stirling engine or small organic rankine cycle system. another very interesting thing to research: Tesla turbines. i think they have a lot of potential for home built systems.

noxx - 18-1-2008 at 23:32

Ha thats sad... It would be great if they had a good yield...
One could use their electricity to electrolyse water to make HHO gas and then use the gas to generate more electricity !

But it seems impractical...

Thanks for the answers everyone.