[Stoves] Using a differential windlass as motive power for stove

[andrew]

On Sunday 10 December 2006 12:49, Steve Taylor wrote:
> Do we have cheap candidate metals for the thermopile idea, in your
> style, possibly with my water jacket ? If we can use strips,
> rather than Belleville washers, we can save a lot of material, and
> shut a lot of people up.


Aluminum 	3.5 	  	Gold 	6.5 	  	Rhodium 	6.0
Antimony 	         47 	  	Iron 	19 	  	Selenium 	900
Bismuth 	        -72 	  	Lead 	4.0 	  	Silicon 	440
Cadmium  	7.5 	  	Mercury 	0.60 	  	Silver 	6.5
Carbon 	        3.0 	  	Nichrome 	25 	  	Sodium 	-2.0
Constantan 	-35 	  	Nickel 	-15 	  	Tantalum 	4.5
Copper 	        6.5 	  	Platinum 	0 	  	Tellurium 	500
Germanium 	300 	  	Potassium 	-9.0 	  	Tungsten 	7.5

Above is a table of Seebeck coefficients lifted from:

<http://www.efunda.com/DesignStandards/sensors/thermocouples/thmcple_theory.cfm>

The coefficient is mV/DegC |( though I thought micro rather than 
milli volts were more likely).

So the idea is to choose two metals that are far apart. I don't know 
what other characteristics these metals have but I would have 
thought high electrical conductivity would need to be one.

You will see silicon and bismuth appear to be good. For common metals 
iron and nickel seemed a useful pair. Though corrosion of the iron 
may be an issue. Nichrome is readily available as wire.

The washers seemed to fill the need for closing the thermocouple from 
the cold to hot side plus enabling some compression to form a good 
junction without welding, though welding does seem best.

AJH