Electric energy from thermocouples

[Brainiac]

If I were to make millions of contacts or "spots" on a long wire,and connect them in series(or parallel?),then make half of them warmer than the other half,would I get any useful voltage,i.e. electrical energy I could use to power something?

 

[Dale]

Yes, this is known as a thermoelectric generator. They are rather inefficient heat engines, but they do have some practical applications.

 

[Danger]

I always make allowances for the inclusion of thermionic converters in appropriate circumstances for things that I design. For instance, wrapping the hottest parts of an automotive exhaust system in a blanket of them can provide enough juice to run your radio and GPS at least, and perhaps more. It really wouldn't be cost effective, but it might save a small fraction of a mpg via decreased electromagnetic drag upon your alternator.
It's a fun exercise, but not worth doing in reality.

 

[Brainiac]

Yes, this is known as a thermoelectric generator. They are rather inefficient heat engines, but they do have some practical applications.

But why is it inefficient? Also one little question: would I need to connect them in series or parallel?

 

[Danger]

would I need to connect them in series or parallel?

That depends upon whether you require more voltage or more current.

 

[Dale]

But why is it inefficient?

The efficiency is limited by the thermal conductivity, the electrical conductivity and the Seebeck coefficient of the materials involved:

http://en.wikipedia.org/wiki/Thermoelectric_materials#Device_efficiency