> A thermoelectric generator is little more than two dissimilar metals
> twisted together with one end heated and the other end cooled.

> The question is how to do it in such a way that useful power can be
> extracted.

There is a superb online tutorial and calculator on this - possibly on 
the site I referenced within the last day.
Ah

    http://www.ferrotec.com/usa/thermoelectric/ref/3ref13.htm

There's no magic, but the engineering gets hard.

Basically you "simply" have to provide a suitably low resistance path 
through the TE cell elements to give it an adequately low internal 
resistance.
TEG = thermoelectric generator = N cells in series or series parallel.

Voltage is calculated from Seebeck effect coefficients for the 
substances used.
TEG resistance is the sum of N cells in series used to get realistic 
voltages.
Resistance is N x cell resistances plus n-1 bridges between them plus 
connecting wires.
voltage increases with temperature differential but the substances 
with best Seebeck coefficient tend to melt /die at lower temperatures.

___________

Optimal design of a multi-couple thermoelectric generator
Heavyish. But useful and interesting.
Answers Byron's question - albeit not as he would have wished it 
answered :-)

        http://www.paper.edu.cn/scholar/download.jsp?file=chenjincan-6

Just to annoy
The thermoelectric generator as an endoreversible Carnot engine

        http://www.iop.org/EJ/abstract/0022-3727/30/3/007


        RM 

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist