[Stoves] Peltier devices and PWM

[steve]

> What Robert Penn Taylor is pointing out is that a device
> designed for cooling, which is what Steve originally
> pointed me to, will not be optimised for generating
> electricity nor withstanding the temperatures normally
> associated with a device purpose made to act as a Thermo
> Electric Generator. Which is fair comment but the point
> was these Peltier devices have become cheaper faster than
> the dedicated Seebeck devices and could be made to work
> running a stove fan. >

The cost/benefit ratio really says that using a TEG is the
best way to go forward - you and I have both played with the
baleful thermoelectric voltage tables to no good effect.
What IS needed is an absolute method of ensuring that the
TEG wouldn't exceed 150C when it is placed in the stove.
Personally, I would put an element that can drop the TEG off
the stove if it reaches its limit - there are solders that
could do that. 

One thing not well addressed is the precision of surface
that a TEG likes to be pressed against - we use them (as
coolers) in the instruments I design in my company, and we
hand lap the heatsinks to make them fit the cell perfectly.
The devices last much longer that way. In common use (and I
look at this as a hobby...) the commercial fridges using
Peltier cells rarely even machine the surfaces, let alone
lap them. I have got an extra 4 degrees of cooling from our
camping fridge by lapping the bits and pieces.

Steve