Thermocouples: Size vs. Number, plus other questions

[Sravoff]

I am working on a solar oven powered thermopile and am trying to figure out how big to make the individual thermocouples. The easiest materials for me to get a hold of will be iron and nickel, Which will give me a seebeck coefficient difference of about 34, so I will need quite a few of these. How does the size of the metals used change the volts/amperes? How does a junction of two one inch-diameter bars differ from a junction of wires? O haven't found anything online that talks about this yet, and would love a pointer in the right direction.

Also, I am planning on using a parabolic solar oven to do the heating work. My design is a stack: ceramic tile, thermopile, ceramic pile, piped water cooled by an evaporation apparatus, a heat sink, then a shade structure on top covered in the same reflective surface as the dish to keep the sun from baking the heat sink. If anyone has suggestions on that aspect of the design, I am all ears.

I am basing the stack off of this design:http://www.tegpower.com/index.html, though I added the water cooling to it. I am unsure how to get my hands on actual semiconductors, thus my iron and nickel plan.

With all of this I hope to run a small fm radio while the sun is out, so I will only need around 9 volts. I am going off the assumption I can connect thermo couples in series/parallel as I could with dc batteries to attain the required voltage, is that correct?

 

[NascentOxygen]

I am working on a solar oven powered thermopile and am trying to figure out how big to make the individual thermocouples. The easiest materials for me to get a hold of will be iron and nickel, Which will give me a seebeck coefficient difference of about 34, so I will need quite a few of these. How does the size of the metals used change the volts/amperes? How does a junction of two one inch-diameter bars differ from a junction of wires?

Physical size makes no difference to the potential developed, as far as I'm aware. But heavier gauge wires will have lower resistance, and if you intend building up a high voltage stack its current capacity will be limited by its resistance.

Thermopiles in domestic use have been tried before. I recall Soviet Russia manufactured a home kerosene lamp with an integrated thermopile, the device providing the peasant with light, heat, and electricity sufficient to power a modest radio. It was not a compact design.

Yes, you connect thermocouples in series for a higher voltage.

 

[Sravoff]

Thanks Nascent, that is what I needed to know.

The teg power link I posted claims to have created a few thermo piles that charge a 12 volt battery off a wood stove. They use n and p semiconductors, where I'll be using nickel/iron junctions so mine won't at all be as efficient, but I think theirs is a step up from the soviet kerosene thermopile.

 

[NascentOxygen]

I'm sure the web will be a good source of hobbyist articles on DIY thermocouples. But I expect "iron" to also be associated with a short life due to corrosion, unless steps are taken to inhibit it. Also, creating reliable metal-metal junctions might be difficult, and a spot-welder may provide the answer; failing that, a powerful crimper. Twisted joins would be good only for a demo model. A high temperature solder would be dicey.

Good luck with your project.

 

[Sravoff]

Yeah, the iron is beginning to look like a problem, I'll have to fix that somehow.

Going from zero to a working thermocouple is proving to be a path fraught with many dead ends--helluvah learning experience though, thanks!